1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719 2720 2721 2722 2723 2724 2725 2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 2793 2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835 2836 2837 2838 2839 2840 2841 2842 2843 2844 2845 2846 2847 2848 2849 2850 2851 2852 2853 2854 2855 2856 2857 2858 2859 2860 2861 2862 2863 2864 2865 2866 2867 2868 2869 2870 2871 2872 2873 2874 2875 2876 2877 2878 2879 2880 2881 2882 2883 2884 2885 2886 2887 2888 2889 2890 2891 2892 2893 2894 2895 2896 2897 2898 2899 2900 2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 2913 2914 2915 2916 2917 2918 2919 2920 2921 2922 2923 2924 2925 2926 2927 2928 2929 2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940 2941 2942 2943 2944 2945 2946 2947 2948 2949 2950 2951 2952 2953 2954 2955 2956 2957 2958 2959 2960 2961 2962 2963 2964 2965 2966 2967 2968 2969 2970 2971 2972 2973 2974 2975 2976 2977 2978 2979 2980 2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007 3008 3009 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 3028 3029 3030 3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049 3050 3051 3052 3053 3054 3055 3056 3057 3058 3059 3060 3061 3062 3063 3064 3065 3066 3067 3068 3069 3070 3071 3072 3073 3074 3075 3076 3077 3078 3079 3080 3081 3082 3083 3084 3085 3086 3087 3088 3089 3090 3091 3092 3093 3094 3095 3096 3097 3098 3099 3100 3101 3102 3103 3104 3105 3106 3107 3108 3109 3110 3111 3112 3113 3114 3115 3116 3117 3118 3119 3120 3121 3122 3123 3124 3125 3126 3127 3128 3129 3130 3131 3132 3133 3134 3135 3136 3137 3138 3139 3140 3141 3142 3143 3144 3145 3146 3147 3148 3149 3150 3151 3152 3153 3154 3155 3156 3157 3158 3159 3160 3161 3162 3163 3164 3165 3166 3167 3168 3169 3170 3171 3172 3173 3174 3175 3176 3177 3178 3179 3180 3181 3182 3183 3184 3185 3186 3187 3188 3189 3190 3191 3192 3193 3194 3195 3196 3197 3198 3199 3200 3201 3202 3203 3204 3205 3206 3207 3208 3209 3210 3211 3212 3213 3214 3215 3216 3217 3218 3219 3220 3221 3222 3223 3224 3225 3226 3227 3228 3229 3230 3231 3232 3233 3234 3235 3236 3237 3238 3239 3240 3241 3242 3243 3244 3245 3246 3247 3248 3249 3250 3251 3252 3253 3254 3255 3256 3257 3258 3259 3260 3261 3262 3263 3264 3265 3266 3267 3268 3269 3270 3271 3272 3273 3274 3275 3276 3277 3278 3279 3280 3281 3282 3283 3284 3285 3286 3287 3288 3289 3290 3291 3292 3293 3294 3295 3296 3297 3298 3299 3300 3301 3302 3303 3304 3305 3306 3307 3308 3309 3310 3311 3312 3313 3314 3315 3316 3317 3318 3319 3320 3321 3322 3323 3324 3325 3326 3327 3328 3329 3330 3331 3332 3333 3334 3335 3336 3337 3338 3339 3340 3341 3342 3343 3344 3345 3346 3347 3348 3349 3350 3351 3352 3353 3354 3355 3356 3357 3358 3359 3360 3361 3362 3363 3364 3365 3366 3367 3368 3369 3370 3371 3372 3373 3374 3375 3376 3377 3378 3379 3380 3381 3382 3383 3384 3385 3386 3387 3388 3389 3390 3391 3392 3393 3394 3395 3396 3397 3398 3399 3400 3401 3402 3403 3404 3405 3406 3407 3408 3409 3410 3411 3412 3413 3414 3415 3416 3417 3418 3419 3420 3421 3422 3423 3424 3425 3426 3427 3428 3429 3430 3431 3432 3433 3434 3435 3436 3437 3438 3439 3440 3441 3442 3443 3444 3445 3446 3447 3448 3449 3450 3451 3452 3453 3454 3455 3456 3457 3458 3459 3460 3461 3462 3463 3464 3465 3466 3467 3468 3469 3470 3471 3472 3473 3474 3475 3476 3477 3478 3479 3480 3481 3482 3483 3484 3485 3486 3487 3488 3489 3490 3491 3492 3493 3494 3495 3496 3497 3498 3499 3500 3501 3502 3503 3504 3505 3506 3507 3508 3509 3510 3511 3512 3513 3514 3515 3516 3517 3518 3519 3520 3521 3522 3523 3524 3525 3526 3527 3528 3529 3530 3531 3532 3533 3534 3535 3536 3537 3538 3539 3540 3541 3542 3543 3544 3545 3546 3547 3548 3549 3550 3551 3552 3553 3554 3555 3556 3557 3558 3559 3560 3561 3562 3563 3564 3565 3566 3567 3568 3569 3570 3571 3572 3573 3574 3575 3576 3577 3578 3579 3580 3581 3582 3583 3584 3585 3586 3587 3588 3589 3590 3591 3592 3593 3594 3595 3596 3597 3598 3599 3600 3601 3602 3603 3604 3605 3606 3607 3608 3609 3610 3611 3612 3613 3614 3615 3616 3617 3618 3619 3620 3621 3622 3623 3624 3625 3626 3627 3628 3629 3630 3631 3632 3633 3634 3635 3636 3637 3638 3639 3640 3641 3642 3643 3644 3645 3646 3647 3648 3649 3650 3651 3652 3653 3654 3655 3656 3657 3658 3659 3660 3661 3662 3663 3664 3665 3666 3667 3668 3669 3670 3671 3672 3673 3674 3675 3676 3677 3678 3679 3680 3681 3682 3683 3684 3685 3686 3687 3688 3689 3690 3691 3692 3693 3694 3695 3696 3697 3698 3699 3700 3701 3702 3703 3704 3705 3706 3707 3708 3709 3710 3711 3712 3713 3714 3715 3716 3717 3718 3719 3720 3721 3722 3723 3724 3725 3726 3727 3728 3729 3730 3731 3732 3733 3734 3735 3736 3737 3738 3739 3740 3741 3742 3743 3744 3745 3746 3747 3748 3749 3750 3751 3752 3753 3754 3755 3756 3757 3758 3759 3760 3761 3762 3763 3764 3765 3766 3767 3768 3769 3770 3771 3772 3773 3774 3775 3776 3777 3778 3779 3780 3781 3782 3783 3784 3785 3786 3787 3788 3789 3790 3791 3792 3793 3794 3795 3796 3797 3798 3799 3800 3801 3802 3803 3804 3805 3806 3807 3808 3809 3810 3811 3812 3813 3814 3815 3816 3817 3818 3819 3820 3821 3822 3823 3824 3825 3826 3827 3828 3829 3830 3831 3832 3833 3834 3835 3836 3837 3838 3839 3840 3841 3842 3843 3844 3845 3846 3847 3848 3849 3850 3851 3852 3853 3854 3855 3856 3857 3858 3859 3860 3861 3862 3863 3864 3865 3866 3867 3868 3869 3870 3871 3872 3873 3874 3875 3876 3877 3878 3879 3880 3881 3882 3883 3884 3885 3886 3887 3888 3889 3890 3891 3892 3893 3894 3895 3896 3897 3898 3899 3900 3901 3902 3903 3904 3905 3906 3907 3908 3909 3910 3911 3912 3913 3914 3915 3916 3917 3918 3919 3920 3921 3922 3923 3924 3925 3926 3927 3928 3929 3930 3931 3932 3933 3934 3935 3936 3937 3938 3939 3940 3941 3942 3943 3944 3945 3946 3947 3948 3949 3950 3951 3952 3953 3954 3955 3956 3957 3958 3959 3960 3961 3962 3963 3964 3965 3966 3967 3968 3969 3970 3971 3972 3973 3974 3975 3976 3977 3978 3979 3980 3981 3982 3983 3984 3985 3986 3987 3988 3989 3990 3991 3992 3993 3994 3995 3996 3997 3998 3999 4000 4001 4002 4003 4004 4005 4006 4007 4008 4009 4010 4011 4012 4013 4014 4015 4016 4017 4018 4019 4020 4021 4022 4023 4024 4025 4026 4027 4028 4029 4030 4031 4032 4033 4034 4035 4036 4037 4038 4039 4040 4041 4042 4043 4044 4045 4046 4047 4048 4049 4050 4051 4052 4053 4054 4055 4056 4057 4058 4059 4060 4061 4062 4063 4064 4065 4066 4067 4068 4069 4070 4071 4072 4073 4074 4075 4076 4077 4078 4079 4080 4081 4082 4083 4084 4085 4086 4087 4088 4089 4090 4091 4092 4093 4094 4095 4096 4097 4098 4099 4100 4101 4102 4103 4104 4105 4106 4107 4108 4109 4110 4111 4112 4113 4114 4115 4116 4117 4118 4119 4120 4121 4122 4123 4124 4125 4126 4127 4128 4129 4130 4131 4132 4133 4134 4135 4136 4137 4138 4139 4140 4141 4142 4143 4144 4145 4146 4147 4148 4149 4150 4151 4152 4153 4154 4155 4156 4157 4158 4159 4160 4161 4162 4163 4164 4165 4166 4167 4168 4169 4170 4171 4172 4173 4174 4175 4176 4177 4178 4179 4180 4181 4182 4183 4184 4185 4186 4187 4188 4189 4190 4191 4192 4193 4194 4195 4196 4197 4198 4199 4200 4201 4202 4203 4204 4205 4206 4207 4208 4209 4210 4211 4212 4213 4214 4215 4216 4217 4218 4219 4220 4221 4222 4223 4224 4225 4226 4227 4228 4229 4230 4231 4232 4233 4234 4235 4236 4237 4238 4239 4240 4241 4242 4243 4244 4245 4246 4247 4248 4249 4250 4251 4252 4253 4254 4255 4256 4257 4258 4259 4260 4261 4262 4263 4264 4265 4266 4267 4268 4269 4270 4271 4272 4273 4274 4275 4276 4277 4278 4279 4280 4281 4282 4283 4284 4285 4286 4287 4288 4289 4290 4291 4292 4293 4294 4295 4296 4297 4298 4299 4300 4301 4302 4303 4304 4305 4306 4307 4308 4309 4310 4311 4312 4313 4314 4315 4316 4317 4318 4319 4320 4321 4322 4323 4324 4325 4326 4327 4328 4329 4330 4331 4332 4333 4334 4335 4336 4337 4338 4339 4340 4341 4342 4343 4344 4345 4346 4347 4348 4349 4350 4351 4352 4353 4354 4355 4356 4357 4358 4359 4360 4361 4362 4363 4364 4365 4366 4367 4368 4369 4370 4371 4372 4373 4374 4375 4376 4377 4378 4379 4380 4381 4382 4383 4384 4385 4386 4387 4388 4389 4390 4391 4392 4393 4394 4395 4396 4397 4398 4399 4400 4401 4402 4403 4404 4405 4406 4407 4408 4409 4410 4411 4412 4413 4414 4415 4416 4417 4418 4419 4420 4421 4422 4423 4424 4425 4426 4427 4428 4429 4430 4431 4432 4433 4434 4435 4436 4437 4438 4439 4440 4441 4442 4443 4444 4445 4446 4447 4448 4449 4450 4451 4452 4453 4454 4455 4456 4457 4458 4459 4460 4461 4462 4463 4464 4465 4466 4467 4468 4469 4470 4471 4472 4473 4474 4475 4476 4477 4478 4479 4480 4481 4482 4483 4484 4485 4486 4487 4488 4489 4490 4491 4492 4493 4494 4495 4496 4497 4498 4499 4500 4501 4502 4503 4504 4505 4506 4507 4508 4509 4510 4511 4512 4513 4514 4515 4516 4517 4518 4519 4520 4521 4522 4523 4524 4525 4526 4527 4528 4529 4530 4531 4532 4533 4534 4535 4536 4537 4538 4539 4540 4541 4542 4543 4544 4545 4546 4547 4548 4549 4550 4551 4552 4553 4554 4555 4556 4557 4558 4559 4560 4561 4562 4563 4564 4565 4566 4567 4568 4569 4570 4571 4572 4573 4574 4575 4576 4577 4578 4579 4580 4581 4582 4583 4584 4585 4586 4587 4588 4589 4590 4591 4592 4593 4594 4595 4596 4597 4598 4599 4600 4601 4602 4603 4604 4605 4606 4607 4608 4609 4610 4611 4612 4613 4614 4615 4616 4617 4618 4619 4620 4621 4622 4623 4624 4625 4626 4627 4628 4629 4630 4631 4632 4633 4634 4635 4636 4637 4638 4639 4640 4641 4642 4643 4644 4645 4646 4647 4648 4649 4650 4651 4652 4653 4654 4655 4656 4657 4658 4659 4660 4661 4662 4663 4664 4665 4666 4667 4668 4669 4670 4671 4672 4673 4674 4675 4676 4677 4678 4679 4680 4681 4682 4683 4684 4685 4686 4687 4688 4689 4690 4691 4692 4693 4694 4695 4696 4697 4698 4699 4700 4701 4702 4703 4704 4705 4706 4707 4708 4709 4710 4711 4712 4713 4714 4715 4716 4717 4718 4719 4720 4721 4722 4723 4724 4725 4726 4727 4728 4729 4730 4731 4732 4733 4734 4735 4736 4737 4738 4739 4740 4741 4742 4743 4744 4745 4746 4747 4748 4749 4750 4751 4752 4753 4754 4755 4756 4757 4758 4759 4760 4761 4762 4763 4764 4765 4766 4767 4768 4769 4770 4771 4772 4773 4774 4775 4776 4777 4778 4779 4780 4781 4782 4783 4784 4785 4786 4787 4788 4789 4790 4791 4792 4793 4794 4795 4796 4797 4798 4799 4800 4801 4802 4803 4804 4805 4806 4807 4808 4809 4810 4811 4812 4813 4814 4815 4816 4817 4818 4819 4820 4821 4822 4823 4824 4825 4826 4827 4828 4829 4830 4831 4832 4833 4834 4835 4836 4837 4838 4839 4840 4841 4842 4843 4844 4845 4846 4847 4848 4849 4850 4851 4852 4853 4854 4855 4856 4857 4858 4859 4860 4861 4862 4863 4864 4865 4866 4867 4868 4869 4870 4871 4872 4873 4874 4875 4876 4877 4878 4879 4880 4881 4882 4883 4884 4885 4886 4887 4888 4889 4890 4891 4892 4893 4894 4895 4896 4897 4898 4899 4900 4901 4902 4903 4904 4905 4906 4907 4908 4909 4910 4911 4912 4913 4914 4915 4916 4917 4918 4919 4920 4921 4922 4923 4924 4925 4926 4927 4928 4929 4930 4931 4932 4933 4934 4935 4936 4937 4938 4939 4940 4941 4942 4943 4944 4945 4946 4947 4948 4949 4950 4951 4952 4953 4954 4955 4956 4957 4958 4959 4960 4961 4962 4963 4964 4965 4966 4967 4968 4969 4970 4971 4972 4973 4974 4975 4976 4977 4978 4979 4980 4981 4982 4983 4984 4985 4986 4987 4988 4989 4990 4991 4992 4993 4994 4995 4996 4997 4998 4999 5000 5001 5002 5003 5004 5005 5006 5007 5008 5009 5010 5011 5012 5013 5014 5015 5016 5017 5018 5019 5020 5021 5022 5023 5024 5025 5026 5027 5028 5029 5030 5031 5032 5033 5034 5035 5036 5037 5038 5039 5040 5041 5042 5043 5044 5045 5046 5047 5048 5049 5050 5051 5052 5053 5054 5055 5056 5057 5058 5059 5060 5061 5062 5063 5064 5065 5066 5067 5068 5069 5070 5071 5072 5073 5074 5075 5076 5077 5078 5079 5080 5081 5082 5083 5084 5085 5086 5087 5088 5089 5090 5091 5092 5093 5094 5095 5096 5097 5098 5099 5100 5101 5102 5103 5104 5105 5106 5107 5108 5109 5110 5111 5112 5113 5114 5115 5116 5117 5118 5119 5120 5121 5122 5123 5124 5125 5126 5127 5128 5129 5130 5131 5132 5133 5134 5135 5136 5137 5138 5139 5140 5141 5142 5143 5144 5145 5146 5147 5148 5149 5150 5151 5152 5153 5154 5155 5156 5157 5158 5159 5160 5161 5162 5163 5164 5165 5166 5167 5168 5169 5170 5171 5172 5173 5174 5175 5176 5177 5178 5179 5180 5181 5182 5183 5184 5185 5186 5187 5188 5189 5190 5191 5192 5193 5194 5195 5196 5197 5198 5199 5200 5201 5202 5203 5204 5205 5206 5207 5208 5209 5210 5211 5212 5213 5214 5215 5216 5217 5218 5219 5220 5221 5222 5223 5224 5225 5226 5227 5228 5229 5230 5231 5232 5233 5234 5235 5236 5237 5238 5239 5240 5241 5242 5243 5244 5245 5246 5247 5248 5249 5250 5251 5252 5253 5254 5255 5256 5257 5258 5259 5260 5261 5262 5263 5264 5265 5266 5267 5268 5269 5270 5271 5272 5273 5274 5275 5276 5277 5278 5279 5280 5281 5282 5283 5284 5285 5286 5287 5288 5289 5290 5291 5292 5293 5294 5295 5296
|
/*****************************************************************************\
* slurm.h - Definitions for all of the Slurm RPCs
*****************************************************************************
* Copyright (C) 2002-2007 The Regents of the University of California.
* Copyright (C) 2008-2010 Lawrence Livermore National Security.
* Portions Copyright (C) 2010-2017 SchedMD LLC <https://www.schedmd.com>.
* Portions Copyright (C) 2012-2013 Los Alamos National Security, LLC.
* Portions Copyright 2013 Hewlett Packard Enterprise Development LP
* Produced at Lawrence Livermore National Laboratory (cf, DISCLAIMER).
* Written by Morris Jette <jette1@llnl.gov>, et. al.
* CODE-OCEC-09-009. All rights reserved.
*
* This file is part of Slurm, a resource management program.
* For details, see <https://slurm.schedmd.com/>.
* Please also read the included file: DISCLAIMER.
*
* Slurm is free software; you can redistribute it and/or modify it under
* the terms of the GNU General Public License as published by the Free
* Software Foundation; either version 2 of the License, or (at your option)
* any later version.
*
* In addition, as a special exception, the copyright holders give permission
* to link the code of portions of this program with the OpenSSL library under
* certain conditions as described in each individual source file, and
* distribute linked combinations including the two. You must obey the GNU
* General Public License in all respects for all of the code used other than
* OpenSSL. If you modify file(s) with this exception, you may extend this
* exception to your version of the file(s), but you are not obligated to do
* so. If you do not wish to do so, delete this exception statement from your
* version. If you delete this exception statement from all source files in
* the program, then also delete it here.
*
* Slurm is distributed in the hope that it will be useful, but WITHOUT ANY
* WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
* FOR A PARTICULAR PURPOSE. See the GNU General Public License for more
* details.
*
* You should have received a copy of the GNU General Public License along
* with Slurm; if not, write to the Free Software Foundation, Inc.,
* 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
\*****************************************************************************/
#ifndef _SLURM_H
#define _SLURM_H
/* Number of dimensions the system has */
#define SYSTEM_DIMENSIONS 1
#define HIGHEST_DIMENSIONS 5
#ifdef __cplusplus
extern "C" {
#endif
#include <slurm/slurm_errno.h>
#include <slurm/slurm_version.h>
#include <inttypes.h> /* for uint16_t, uint32_t definitions */
#include <netinet/in.h> /* struct sockaddr_in */
#include <stdbool.h>
#include <stdio.h> /* for FILE definitions */
#include <sys/types.h> /* for uid_t definition */
#include <time.h> /* for time_t definitions */
#include <unistd.h>
/* Slurm Lexicographically-sortable Unique ID */
typedef uint64_t sluid_t;
/* Define slurm_addr_t below to avoid including extraneous slurm headers */
typedef struct sockaddr_storage slurm_addr_t;
#ifndef __slurmdb_cluster_rec_t_defined
# define __slurmdb_cluster_rec_t_defined
typedef struct slurmdb_cluster_rec slurmdb_cluster_rec_t;
#endif
/* Define job_resources_t below
* to avoid including extraneous slurm headers */
#ifndef __job_resources_t_defined
# define __job_resources_t_defined /* Opaque data for select plugins */
typedef struct job_resources job_resources_t;
#endif
/* Define select_jobinfo_t, select_nodeinfo_t below
* to avoid including extraneous slurm headers */
#ifndef __select_jobinfo_t_defined
# define __select_jobinfo_t_defined /* Opaque data for select plugins */
typedef struct select_jobinfo select_jobinfo_t; /* for BlueGene */
typedef struct select_nodeinfo select_nodeinfo_t; /* for BlueGene */
#endif
/* Define jobacctinfo_t below to avoid including extraneous slurm headers */
#ifndef __jobacctinfo_t_defined
# define __jobacctinfo_t_defined
typedef struct jobacctinfo jobacctinfo_t; /* opaque data type */
#endif
/* Define allocation_msg_thread_t below to avoid including extraneous
* slurm headers */
#ifndef __allocation_msg_thread_t_defined
# define __allocation_msg_thread_t_defined
typedef struct allocation_msg_thread allocation_msg_thread_t;
#endif
#ifndef __sbcast_cred_t_defined
# define __sbcast_cred_t_defined
typedef struct sbcast_cred sbcast_cred_t; /* opaque data type */
#endif
/*****************************************************************************\
* DEFINITIONS FOR POSIX VALUES
\*****************************************************************************/
#ifndef HOST_NAME_MAX
#define HOST_NAME_MAX 64
#endif
/*****************************************************************************\
* DEFINITIONS FOR INPUT VALUES
\*****************************************************************************/
/* INFINITE is used to identify unlimited configurations, */
/* eg. the maximum count of nodes any job may use in some partition */
#define INFINITE8 (0xff)
#define INFINITE16 (0xffff)
#define INFINITE (0xffffffff)
#define INFINITE64 (0xffffffffffffffff)
#define NO_VAL8 (0xfe)
#define NO_VAL16 (0xfffe)
#define NO_VAL (0xfffffffe)
#define NO_VAL64 (0xfffffffffffffffe)
#define NO_CONSUME_VAL64 (0xfffffffffffffffd)
#define MAX_TASKS_PER_NODE 512
#define MAX_JOB_ID (0x03FFFFFF) /* bits 0-25 */
#define MAX_FED_JOB_ID (0xfffffffd) /* NO_VAL - 1 */
#define MAX_HET_JOB_COMPONENTS 128
#define MAX_FED_CLUSTERS 63
#define MAX_JOB_SIZE_BITMAP 16384
#define MAX_SLURM_NODES 65536 /* (64*1024) hosts */
/*
* Max normal step id leaving a few for special steps like the batch and extern
* steps
*/
#define SLURM_MAX_NORMAL_STEP_ID (0xfffffff0)
/* Job step ID of pending step */
#define SLURM_PENDING_STEP (0xfffffffd)
/* Job step ID of external process container */
#define SLURM_EXTERN_CONT (0xfffffffc)
/* Job step ID of batch scripts */
#define SLURM_BATCH_SCRIPT (0xfffffffb)
/* Job step ID for the interactive step (if used) */
#define SLURM_INTERACTIVE_STEP (0xfffffffa)
/* How many seconds to wait after eio_signal_shutdown() is called before
* terminating the job and abandoning any I/O remaining to be processed.
*/
#define DEFAULT_EIO_SHUTDOWN_WAIT 60
/*
* SLURM_ID_HASH
* Description:
* Creates a hash of a Slurm JOBID and STEPID
* The JOB STEP ID is in the top 32 bits of the hash with the job id occupying
* the lower 32 bits.
*
* IN _jobid -- SLURM's JOB ID (uint32_t)
* IN _stepid -- SLURM's JOB STEP ID (uint32_t)
* RET id_hash -- (uint64_t)
*/
#define SLURM_ID_HASH(_jobid, _stepid) \
(uint64_t)(((uint64_t)_stepid << 32) + _jobid)
#define SLURM_ID_HASH_JOB_ID(hash_id) (uint32_t)(hash_id & 0x00000000FFFFFFFF)
#define SLURM_ID_HASH_STEP_ID(hash_id) (uint32_t)(hash_id >> 32)
/*
* Convert a hash ID to its legacy (pre-17.11) equivalent
* Used for backward compatibility for Cray PMI
*/
#define SLURM_ID_HASH_LEGACY(hash_id) \
((hash_id >> 32) * 10000000000 + (hash_id & 0x00000000FFFFFFFF))
/* Slurm hash definition to be used for various purposes */
typedef struct {
unsigned char type;
unsigned char hash[32];
} slurm_hash_t;
/*
* Bit definitions when setting flags
*
* SLURM_BIT(0) 0x0000000000000001
* SLURM_BIT(1) 0x0000000000000002
* SLURM_BIT(2) 0x0000000000000004
* SLURM_BIT(3) 0x0000000000000008
* SLURM_BIT(4) 0x0000000000000010
* SLURM_BIT(5) 0x0000000000000020
* SLURM_BIT(6) 0x0000000000000040
* SLURM_BIT(7) 0x0000000000000080
* SLURM_BIT(8) 0x0000000000000100
* SLURM_BIT(9) 0x0000000000000200
* SLURM_BIT(10) 0x0000000000000400
* SLURM_BIT(11) 0x0000000000000800
* SLURM_BIT(12) 0x0000000000001000
* SLURM_BIT(13) 0x0000000000002000
* SLURM_BIT(14) 0x0000000000004000
* SLURM_BIT(15) 0x0000000000008000
* SLURM_BIT(16) 0x0000000000010000
* SLURM_BIT(17) 0x0000000000020000
* SLURM_BIT(18) 0x0000000000040000
* SLURM_BIT(19) 0x0000000000080000
* SLURM_BIT(20) 0x0000000000100000
* SLURM_BIT(21) 0x0000000000200000
* SLURM_BIT(22) 0x0000000000400000
* SLURM_BIT(23) 0x0000000000800000
* SLURM_BIT(24) 0x0000000001000000
* SLURM_BIT(25) 0x0000000002000000
* SLURM_BIT(26) 0x0000000004000000
* SLURM_BIT(27) 0x0000000008000000
* SLURM_BIT(28) 0x0000000010000000
* SLURM_BIT(29) 0x0000000020000000
* SLURM_BIT(30) 0x0000000040000000
* SLURM_BIT(31) 0x0000000080000000
* SLURM_BIT(32) 0x0000000100000000
* SLURM_BIT(33) 0x0000000200000000
* SLURM_BIT(34) 0x0000000400000000
* SLURM_BIT(35) 0x0000000800000000
* SLURM_BIT(36) 0x0000001000000000
* SLURM_BIT(37) 0x0000002000000000
* SLURM_BIT(38) 0x0000004000000000
* SLURM_BIT(39) 0x0000008000000000
* SLURM_BIT(40) 0x0000010000000000
* SLURM_BIT(41) 0x0000020000000000
* SLURM_BIT(42) 0x0000040000000000
* SLURM_BIT(43) 0x0000080000000000
* SLURM_BIT(44) 0x0000100000000000
* SLURM_BIT(45) 0x0000200000000000
* SLURM_BIT(46) 0x0000400000000000
* SLURM_BIT(47) 0x0000800000000000
* SLURM_BIT(48) 0x0001000000000000
* SLURM_BIT(49) 0x0002000000000000
* SLURM_BIT(50) 0x0004000000000000
* SLURM_BIT(51) 0x0008000000000000
* SLURM_BIT(52) 0x0010000000000000
* SLURM_BIT(53) 0x0020000000000000
* SLURM_BIT(54) 0x0040000000000000
* SLURM_BIT(55) 0x0080000000000000
* SLURM_BIT(56) 0x0100000000000000
* SLURM_BIT(57) 0x0200000000000000
* SLURM_BIT(58) 0x0400000000000000
* SLURM_BIT(59) 0x0800000000000000
* SLURM_BIT(60) 0x1000000000000000
* SLURM_BIT(61) 0x2000000000000000
* SLURM_BIT(62) 0x4000000000000000
* SLURM_BIT(63) 0x8000000000000000
*/
#define SLURM_BIT(offset) ((uint64_t)1 << offset)
/* last entry must be JOB_END, keep in sync with job_state_string and
* job_state_string_compact. values may be ORed with JOB_STATE_FLAGS
* below. */
enum job_states {
JOB_PENDING, /* queued waiting for initiation */
JOB_RUNNING, /* allocated resources and executing */
JOB_SUSPENDED, /* allocated resources, execution suspended */
JOB_COMPLETE, /* completed execution successfully */
JOB_CANCELLED, /* cancelled by user */
JOB_FAILED, /* completed execution unsuccessfully */
JOB_TIMEOUT, /* terminated on reaching time limit */
JOB_NODE_FAIL, /* terminated on node failure */
JOB_PREEMPTED, /* terminated due to preemption */
JOB_BOOT_FAIL, /* terminated due to node boot failure */
JOB_DEADLINE, /* terminated on deadline */
JOB_OOM, /* experienced out of memory error */
JOB_END /* not a real state, last entry in table */
};
#define JOB_STATE_BASE 0x000000ff /* Used for job_states above */
#define JOB_STATE_FLAGS 0xffffff00 /* Used for state flags below */
/* SLURM_BIT(0-7) are already taken with base job_states above */
#define JOB_LAUNCH_FAILED SLURM_BIT(8)
/* was JOB_UPDATE_DB SLURM_BIT(9), removed v24.11 */
#define JOB_REQUEUE SLURM_BIT(10) /* Requeue job in completing state */
#define JOB_REQUEUE_HOLD SLURM_BIT(11) /* Requeue any job in hold */
#define JOB_SPECIAL_EXIT SLURM_BIT(12) /* Requeue an exit job in hold */
#define JOB_RESIZING SLURM_BIT(13) /* Size of job about to change, flag set
* before calling accounting functions
* immediately before job changes size
*/
#define JOB_CONFIGURING SLURM_BIT(14) /* Allocated nodes booting */
#define JOB_COMPLETING SLURM_BIT(15) /* Waiting for epilog completion */
#define JOB_STOPPED SLURM_BIT(16) /* Job is stopped state (holding
resources, but sent SIGSTOP */
#define JOB_RECONFIG_FAIL SLURM_BIT(17) /* Node configuration for job failed,
not job state, just job requeue
flag */
#define JOB_POWER_UP_NODE SLURM_BIT(18) /* Allocated powered down nodes,
* waiting for reboot */
#define JOB_REVOKED SLURM_BIT(19) /* Sibling job revoked */
#define JOB_REQUEUE_FED SLURM_BIT(20) /* Job being requeued by federation */
#define JOB_RESV_DEL_HOLD SLURM_BIT(21) /* Job is hold */
#define JOB_SIGNALING SLURM_BIT(22) /* Outgoing signal is pending */
#define JOB_STAGE_OUT SLURM_BIT(23) /* Staging out data (burst buffer) */
#define READY_JOB_FATAL -2 /* fatal error */
#define READY_JOB_ERROR -1 /* ordinary error */
#define READY_NODE_STATE SLURM_BIT(0) /* job's node's are ready */
#define READY_JOB_STATE SLURM_BIT(1) /* job is ready to execute */
#define READY_PROLOG_STATE SLURM_BIT(2) /* PrologSlurmctld is done */
#define MAIL_JOB_BEGIN SLURM_BIT(0) /* Notify when job begins */
#define MAIL_JOB_END SLURM_BIT(1) /* Notify when job ends */
#define MAIL_JOB_FAIL SLURM_BIT(2) /* Notify if job fails */
#define MAIL_JOB_REQUEUE SLURM_BIT(3) /* Notify if job requeued */
#define MAIL_JOB_TIME100 SLURM_BIT(4) /* Notify on reaching 100% of time
* limit */
#define MAIL_JOB_TIME90 SLURM_BIT(5) /* Notify on reaching 90% of time
* limit */
#define MAIL_JOB_TIME80 SLURM_BIT(6) /* Notify on reaching 80% of time
* limit */
#define MAIL_JOB_TIME50 SLURM_BIT(7) /* Notify on reaching 50% of time
* limit */
#define MAIL_JOB_STAGE_OUT SLURM_BIT(8) /* Notify on completion of burst
* buffer stage out */
#define MAIL_ARRAY_TASKS SLURM_BIT(9) /* Send emails for each array task */
#define MAIL_INVALID_DEPEND SLURM_BIT(10) /* Notify on job invalid dependency */
/* Do not fatal on unrecognized input */
#define PARSE_FLAGS_IGNORE_NEW SLURM_BIT(0)
/* Check 0600 permissions for slurmdbd.conf included files */
#define PARSE_FLAGS_CHECK_PERMISSIONS SLURM_BIT(1)
/* Only parse Include directives and ignore rest of parsing */
#define PARSE_FLAGS_INCLUDE_ONLY SLURM_BIT(2)
/*
* job_array_struct_t array_flags definitions. ARRAY_TASK_REQUEUED could be
* substituted in the future to tot_requeued_tasks member in the struct, which
* would provide a more accurated array statistic.
*/
#define ARRAY_TASK_REQUEUED 0x0001 /* At least one task was requeued. */
#define NICE_OFFSET 0x80000000 /* offset for job's nice value */
/* Reason for job to be pending rather than executing or reason for job
* failure. If multiple reasons exists, only one is given for the sake of
* system efficiency */
enum job_state_reason {
/* Reasons for job to be pending */
WAIT_NO_REASON = 0, /* not set or job not pending */
WAIT_PRIORITY, /* higher priority jobs exist */
WAIT_DEPENDENCY, /* dependent job has not completed */
WAIT_RESOURCES, /* required resources not available */
WAIT_PART_NODE_LIMIT, /* request exceeds partition node limit */
WAIT_PART_TIME_LIMIT, /* request exceeds partition time limit */
WAIT_PART_DOWN, /* requested partition is down */
WAIT_PART_INACTIVE, /* requested partition is inactive */
WAIT_HELD, /* job is held by administrator */
WAIT_TIME, /* job waiting for specific begin time */
WAIT_LICENSES, /* job is waiting for licenses */
WAIT_ASSOC_JOB_LIMIT, /* user/bank job limit reached */
WAIT_ASSOC_RESOURCE_LIMIT,/* user/bank resource limit reached */
WAIT_ASSOC_TIME_LIMIT, /* user/bank time limit reached */
WAIT_RESERVATION, /* reservation not available */
WAIT_NODE_NOT_AVAIL, /* required node is DOWN or DRAINED */
WAIT_HELD_USER, /* job is held by user */
WAIT_FRONT_END, /* Front end nodes are DOWN */
FAIL_DEFER, /* individual submit time sched deferred */
FAIL_DOWN_PARTITION, /* partition for job is DOWN */
FAIL_DOWN_NODE, /* some node in the allocation failed */
FAIL_BAD_CONSTRAINTS, /* constraints can not be satisfied */
FAIL_SYSTEM, /* slurm system failure */
FAIL_LAUNCH, /* unable to launch job */
FAIL_EXIT_CODE, /* exit code was non-zero */
FAIL_TIMEOUT, /* reached end of time limit */
FAIL_INACTIVE_LIMIT, /* reached slurm InactiveLimit */
FAIL_ACCOUNT, /* invalid account */
FAIL_QOS, /* invalid QOS */
WAIT_QOS_THRES, /* required QOS threshold has been breached */
WAIT_QOS_JOB_LIMIT, /* QOS job limit reached */
WAIT_QOS_RESOURCE_LIMIT,/* QOS resource limit reached */
WAIT_QOS_TIME_LIMIT, /* QOS time limit reached */
FAIL_SIGNAL, /* raised a signal that caused it to exit */
DEFUNCT_WAIT_34, /* free for reuse */
WAIT_CLEANING, /* If a job is requeued and it is
* still cleaning up from the last run. */
WAIT_PROLOG, /* Prolog is running */
WAIT_QOS, /* QOS not allowed */
WAIT_ACCOUNT, /* Account not allowed */
WAIT_DEP_INVALID, /* Dependency condition invalid or never
* satisfied
*/
WAIT_QOS_GRP_CPU, /* QOS GrpTRES exceeded (CPU) */
WAIT_QOS_GRP_CPU_MIN, /* QOS GrpTRESMins exceeded (CPU) */
WAIT_QOS_GRP_CPU_RUN_MIN, /* QOS GrpTRESRunMins exceeded (CPU) */
WAIT_QOS_GRP_JOB, /* QOS GrpJobs exceeded */
WAIT_QOS_GRP_MEM, /* QOS GrpTRES exceeded (Memory) */
WAIT_QOS_GRP_NODE, /* QOS GrpTRES exceeded (Node) */
WAIT_QOS_GRP_SUB_JOB, /* QOS GrpSubmitJobs exceeded */
WAIT_QOS_GRP_WALL, /* QOS GrpWall exceeded */
WAIT_QOS_MAX_CPU_PER_JOB, /* QOS MaxTRESPerJob exceeded (CPU) */
WAIT_QOS_MAX_CPU_MINS_PER_JOB,/* QOS MaxTRESMinsPerJob exceeded (CPU) */
WAIT_QOS_MAX_NODE_PER_JOB, /* QOS MaxTRESPerJob exceeded (Node) */
WAIT_QOS_MAX_WALL_PER_JOB, /* QOS MaxWallDurationPerJob exceeded */
WAIT_QOS_MAX_CPU_PER_USER, /* QOS MaxTRESPerUser exceeded (CPU) */
WAIT_QOS_MAX_JOB_PER_USER, /* QOS MaxJobsPerUser exceeded */
WAIT_QOS_MAX_NODE_PER_USER, /* QOS MaxTRESPerUser exceeded (Node) */
WAIT_QOS_MAX_SUB_JOB, /* QOS MaxSubmitJobsPerUser exceeded */
WAIT_QOS_MIN_CPU, /* QOS MinTRESPerJob not reached (CPU) */
WAIT_ASSOC_GRP_CPU, /* ASSOC GrpTRES exceeded (CPU) */
WAIT_ASSOC_GRP_CPU_MIN, /* ASSOC GrpTRESMins exceeded (CPU) */
WAIT_ASSOC_GRP_CPU_RUN_MIN, /* ASSOC GrpTRESRunMins exceeded (CPU) */
WAIT_ASSOC_GRP_JOB, /* ASSOC GrpJobs exceeded */
WAIT_ASSOC_GRP_MEM, /* ASSOC GrpTRES exceeded (Memory) */
WAIT_ASSOC_GRP_NODE, /* ASSOC GrpTRES exceeded (Node) */
WAIT_ASSOC_GRP_SUB_JOB, /* ASSOC GrpSubmitJobs exceeded */
WAIT_ASSOC_GRP_WALL, /* ASSOC GrpWall exceeded */
WAIT_ASSOC_MAX_JOBS, /* ASSOC MaxJobs exceeded */
WAIT_ASSOC_MAX_CPU_PER_JOB, /* ASSOC MaxTRESPerJob exceeded (CPU) */
WAIT_ASSOC_MAX_CPU_MINS_PER_JOB,/* ASSOC MaxTRESMinsPerJob
* exceeded (CPU) */
WAIT_ASSOC_MAX_NODE_PER_JOB, /* ASSOC MaxTRESPerJob exceeded (NODE) */
WAIT_ASSOC_MAX_WALL_PER_JOB, /* ASSOC MaxWallDurationPerJob
* exceeded */
WAIT_ASSOC_MAX_SUB_JOB, /* ASSOC MaxSubmitJobsPerUser exceeded */
WAIT_MAX_REQUEUE, /* MAX_BATCH_REQUEUE reached */
WAIT_ARRAY_TASK_LIMIT, /* job array running task limit */
WAIT_BURST_BUFFER_RESOURCE, /* Burst buffer resources */
WAIT_BURST_BUFFER_STAGING, /* Burst buffer file stage-in */
FAIL_BURST_BUFFER_OP, /* Burst buffer operation failure */
DEFUNCT_WAIT_76, /* free for reuse */
DEFUNCT_WAIT_77, /* free for reuse */
WAIT_ASSOC_GRP_UNK, /* ASSOC GrpTRES exceeded
* (Unknown) */
WAIT_ASSOC_GRP_UNK_MIN, /* ASSOC GrpTRESMins exceeded
* (Unknown) */
WAIT_ASSOC_GRP_UNK_RUN_MIN, /* ASSOC GrpTRESRunMins exceeded
* (Unknown) */
WAIT_ASSOC_MAX_UNK_PER_JOB, /* ASSOC MaxTRESPerJob exceeded
* (Unknown) */
WAIT_ASSOC_MAX_UNK_PER_NODE, /* ASSOC MaxTRESPerNode exceeded
* (Unknown) */
WAIT_ASSOC_MAX_UNK_MINS_PER_JOB,/* ASSOC MaxTRESMinsPerJob
* exceeded (Unknown) */
WAIT_ASSOC_MAX_CPU_PER_NODE, /* ASSOC MaxTRESPerNode exceeded (CPU) */
WAIT_ASSOC_GRP_MEM_MIN, /* ASSOC GrpTRESMins exceeded
* (Memory) */
WAIT_ASSOC_GRP_MEM_RUN_MIN, /* ASSOC GrpTRESRunMins exceeded
* (Memory) */
WAIT_ASSOC_MAX_MEM_PER_JOB, /* ASSOC MaxTRESPerJob exceeded (Memory) */
WAIT_ASSOC_MAX_MEM_PER_NODE, /* ASSOC MaxTRESPerNode exceeded (CPU) */
WAIT_ASSOC_MAX_MEM_MINS_PER_JOB,/* ASSOC MaxTRESMinsPerJob
* exceeded (Memory) */
WAIT_ASSOC_GRP_NODE_MIN, /* ASSOC GrpTRESMins exceeded (Node) */
WAIT_ASSOC_GRP_NODE_RUN_MIN, /* ASSOC GrpTRESRunMins exceeded (Node) */
WAIT_ASSOC_MAX_NODE_MINS_PER_JOB,/* ASSOC MaxTRESMinsPerJob
* exceeded (Node) */
WAIT_ASSOC_GRP_ENERGY, /* ASSOC GrpTRES exceeded
* (Energy) */
WAIT_ASSOC_GRP_ENERGY_MIN, /* ASSOC GrpTRESMins exceeded
* (Energy) */
WAIT_ASSOC_GRP_ENERGY_RUN_MIN, /* ASSOC GrpTRESRunMins exceeded
* (Energy) */
WAIT_ASSOC_MAX_ENERGY_PER_JOB, /* ASSOC MaxTRESPerJob exceeded
* (Energy) */
WAIT_ASSOC_MAX_ENERGY_PER_NODE, /* ASSOC MaxTRESPerNode
* exceeded (Energy) */
WAIT_ASSOC_MAX_ENERGY_MINS_PER_JOB,/* ASSOC MaxTRESMinsPerJob
* exceeded (Energy) */
WAIT_ASSOC_GRP_GRES, /* ASSOC GrpTRES exceeded (GRES) */
WAIT_ASSOC_GRP_GRES_MIN, /* ASSOC GrpTRESMins exceeded (GRES) */
WAIT_ASSOC_GRP_GRES_RUN_MIN, /* ASSOC GrpTRESRunMins exceeded (GRES) */
WAIT_ASSOC_MAX_GRES_PER_JOB, /* ASSOC MaxTRESPerJob exceeded (GRES) */
WAIT_ASSOC_MAX_GRES_PER_NODE, /* ASSOC MaxTRESPerNode exceeded (GRES) */
WAIT_ASSOC_MAX_GRES_MINS_PER_JOB,/* ASSOC MaxTRESMinsPerJob
* exceeded (GRES) */
WAIT_ASSOC_GRP_LIC, /* ASSOC GrpTRES exceeded
* (license) */
WAIT_ASSOC_GRP_LIC_MIN, /* ASSOC GrpTRESMins exceeded
* (license) */
WAIT_ASSOC_GRP_LIC_RUN_MIN, /* ASSOC GrpTRESRunMins exceeded
* (license) */
WAIT_ASSOC_MAX_LIC_PER_JOB, /* ASSOC MaxTRESPerJob exceeded
* (license) */
WAIT_ASSOC_MAX_LIC_MINS_PER_JOB,/* ASSOC MaxTRESMinsPerJob exceeded
* (license) */
WAIT_ASSOC_GRP_BB, /* ASSOC GrpTRES exceeded
* (burst buffer) */
WAIT_ASSOC_GRP_BB_MIN, /* ASSOC GrpTRESMins exceeded
* (burst buffer) */
WAIT_ASSOC_GRP_BB_RUN_MIN, /* ASSOC GrpTRESRunMins exceeded
* (burst buffer) */
WAIT_ASSOC_MAX_BB_PER_JOB, /* ASSOC MaxTRESPerJob exceeded
* (burst buffer) */
WAIT_ASSOC_MAX_BB_PER_NODE, /* ASSOC MaxTRESPerNode exceeded
* (burst buffer) */
WAIT_ASSOC_MAX_BB_MINS_PER_JOB,/* ASSOC MaxTRESMinsPerJob exceeded
* (burst buffer) */
WAIT_QOS_GRP_UNK, /* QOS GrpTRES exceeded (Unknown) */
WAIT_QOS_GRP_UNK_MIN, /* QOS GrpTRESMins exceeded (Unknown) */
WAIT_QOS_GRP_UNK_RUN_MIN, /* QOS GrpTRESRunMins exceeded (Unknown) */
WAIT_QOS_MAX_UNK_PER_JOB, /* QOS MaxTRESPerJob exceeded (Unknown) */
WAIT_QOS_MAX_UNK_PER_NODE, /* QOS MaxTRESPerNode exceeded (Unknown) */
WAIT_QOS_MAX_UNK_PER_USER, /* QOS MaxTRESPerUser exceeded (Unknown) */
WAIT_QOS_MAX_UNK_MINS_PER_JOB,/* QOS MaxTRESMinsPerJob
* exceeded (Unknown) */
WAIT_QOS_MIN_UNK, /* QOS MinTRESPerJob exceeded (Unknown) */
WAIT_QOS_MAX_CPU_PER_NODE, /* QOS MaxTRESPerNode exceeded (CPU) */
WAIT_QOS_GRP_MEM_MIN, /* QOS GrpTRESMins exceeded
* (Memory) */
WAIT_QOS_GRP_MEM_RUN_MIN, /* QOS GrpTRESRunMins exceeded
* (Memory) */
WAIT_QOS_MAX_MEM_MINS_PER_JOB,/* QOS MaxTRESMinsPerJob
* exceeded (Memory) */
WAIT_QOS_MAX_MEM_PER_JOB, /* QOS MaxTRESPerJob exceeded (CPU) */
WAIT_QOS_MAX_MEM_PER_NODE, /* QOS MaxTRESPerNode exceeded (MEM) */
WAIT_QOS_MAX_MEM_PER_USER, /* QOS MaxTRESPerUser exceeded (CPU) */
WAIT_QOS_MIN_MEM, /* QOS MinTRESPerJob not reached (Memory) */
WAIT_QOS_GRP_ENERGY, /* QOS GrpTRES exceeded (Energy) */
WAIT_QOS_GRP_ENERGY_MIN, /* QOS GrpTRESMins exceeded (Energy) */
WAIT_QOS_GRP_ENERGY_RUN_MIN, /* QOS GrpTRESRunMins exceeded (Energy) */
WAIT_QOS_MAX_ENERGY_PER_JOB, /* QOS MaxTRESPerJob exceeded (Energy) */
WAIT_QOS_MAX_ENERGY_PER_NODE,/* QOS MaxTRESPerNode exceeded (Energy) */
WAIT_QOS_MAX_ENERGY_PER_USER,/* QOS MaxTRESPerUser exceeded (Energy) */
WAIT_QOS_MAX_ENERGY_MINS_PER_JOB,/* QOS MaxTRESMinsPerJob
* exceeded (Energy) */
WAIT_QOS_MIN_ENERGY, /* QOS MinTRESPerJob not reached (Energy) */
WAIT_QOS_GRP_NODE_MIN, /* QOS GrpTRESMins exceeded (Node) */
WAIT_QOS_GRP_NODE_RUN_MIN, /* QOS GrpTRESRunMins exceeded (Node) */
WAIT_QOS_MAX_NODE_MINS_PER_JOB, /* QOS MaxTRESMinsPerJob
* exceeded (Node) */
WAIT_QOS_MIN_NODE, /* QOS MinTRESPerJob not reached (Node) */
WAIT_QOS_GRP_GRES, /* QOS GrpTRES exceeded (GRES) */
WAIT_QOS_GRP_GRES_MIN, /* QOS GrpTRESMins exceeded (GRES) */
WAIT_QOS_GRP_GRES_RUN_MIN, /* QOS GrpTRESRunMins exceeded (GRES) */
WAIT_QOS_MAX_GRES_PER_JOB, /* QOS MaxTRESPerJob exceeded (GRES) */
WAIT_QOS_MAX_GRES_PER_NODE, /* QOS MaxTRESPerNode exceeded (GRES) */
WAIT_QOS_MAX_GRES_PER_USER, /* QOS MaxTRESPerUser exceeded
* (GRES) */
WAIT_QOS_MAX_GRES_MINS_PER_JOB,/* QOS MaxTRESMinsPerJob
* exceeded (GRES) */
WAIT_QOS_MIN_GRES, /* QOS MinTRESPerJob not reached (CPU) */
WAIT_QOS_GRP_LIC, /* QOS GrpTRES exceeded (license) */
WAIT_QOS_GRP_LIC_MIN, /* QOS GrpTRESMins exceeded (license) */
WAIT_QOS_GRP_LIC_RUN_MIN, /* QOS GrpTRESRunMins exceeded (license) */
WAIT_QOS_MAX_LIC_PER_JOB, /* QOS MaxTRESPerJob exceeded (license) */
WAIT_QOS_MAX_LIC_PER_USER, /* QOS MaxTRESPerUser exceeded (license) */
WAIT_QOS_MAX_LIC_MINS_PER_JOB,/* QOS MaxTRESMinsPerJob exceeded
* (license) */
WAIT_QOS_MIN_LIC, /* QOS MinTRESPerJob not reached
* (license) */
WAIT_QOS_GRP_BB, /* QOS GrpTRES exceeded
* (burst buffer) */
WAIT_QOS_GRP_BB_MIN, /* QOS GrpTRESMins exceeded
* (burst buffer) */
WAIT_QOS_GRP_BB_RUN_MIN, /* QOS GrpTRESRunMins exceeded
* (burst buffer) */
WAIT_QOS_MAX_BB_PER_JOB, /* QOS MaxTRESPerJob exceeded
* (burst buffer) */
WAIT_QOS_MAX_BB_PER_NODE, /* QOS MaxTRESPerNode exceeded
* (burst buffer) */
WAIT_QOS_MAX_BB_PER_USER, /* QOS MaxTRESPerUser exceeded
* (burst buffer) */
WAIT_QOS_MAX_BB_MINS_PER_JOB,/* QOS MaxTRESMinsPerJob exceeded
* (burst buffer) */
WAIT_QOS_MIN_BB, /* QOS MinTRESPerJob not reached
* (burst buffer) */
FAIL_DEADLINE, /* reached deadline */
/* QOS MaxTRESPerAccount */
WAIT_QOS_MAX_BB_PER_ACCT, /* exceeded burst buffer */
WAIT_QOS_MAX_CPU_PER_ACCT, /* exceeded CPUs */
WAIT_QOS_MAX_ENERGY_PER_ACCT, /* exceeded Energy */
WAIT_QOS_MAX_GRES_PER_ACCT, /* exceeded GRES */
WAIT_QOS_MAX_NODE_PER_ACCT, /* exceeded Nodes */
WAIT_QOS_MAX_LIC_PER_ACCT, /* exceeded Licenses */
WAIT_QOS_MAX_MEM_PER_ACCT, /* exceeded Memory */
WAIT_QOS_MAX_UNK_PER_ACCT, /* exceeded Unknown */
/********************/
WAIT_QOS_MAX_JOB_PER_ACCT, /* QOS MaxJobPerAccount exceeded */
WAIT_QOS_MAX_SUB_JOB_PER_ACCT,/* QOS MaxJobSubmitSPerAccount exceeded */
WAIT_PART_CONFIG, /* Generic partition configuration reason */
WAIT_ACCOUNT_POLICY, /* Generic accounting policy reason */
WAIT_FED_JOB_LOCK, /* Can't get fed job lock */
FAIL_OOM, /* Exhausted memory */
WAIT_PN_MEM_LIMIT, /* MaxMemPer[CPU|Node] exceeded */
/* exceeded Billing TRES limits */
WAIT_ASSOC_GRP_BILLING, /* GrpTRES */
WAIT_ASSOC_GRP_BILLING_MIN, /* GrpTRESMins */
WAIT_ASSOC_GRP_BILLING_RUN_MIN, /* GrpTRESRunMins */
WAIT_ASSOC_MAX_BILLING_PER_JOB, /* MaxTRESPerJob */
WAIT_ASSOC_MAX_BILLING_PER_NODE, /* MaxTRESPerNode */
WAIT_ASSOC_MAX_BILLING_MINS_PER_JOB,/* MaxTRESMinsPerJob */
WAIT_QOS_GRP_BILLING, /* GrpTRES */
WAIT_QOS_GRP_BILLING_MIN, /* GrpTRESMins */
WAIT_QOS_GRP_BILLING_RUN_MIN, /* GrpTRESRunMins */
WAIT_QOS_MAX_BILLING_PER_JOB, /* MaxTRESPerJob */
WAIT_QOS_MAX_BILLING_PER_NODE, /* MaxTRESPerNode */
WAIT_QOS_MAX_BILLING_PER_USER, /* MaxTRESPerUser */
WAIT_QOS_MAX_BILLING_MINS_PER_JOB, /* MaxTRESMinsPerJob */
WAIT_QOS_MAX_BILLING_PER_ACCT, /* MaxTRESPerAcct */
WAIT_QOS_MIN_BILLING, /* MinTRESPerJob */
WAIT_RESV_DELETED, /* Reservation was deleted */
WAIT_RESV_INVALID,
FAIL_CONSTRAINTS, /* Constraints cannot currently be satisfied */
WAIT_QOS_MAX_BB_RUN_MINS_PER_ACCT,/* QOS MaxTRESRunMinsPerAccount
* exceeded (burst buffer) */
WAIT_QOS_MAX_BILLING_RUN_MINS_PER_ACCT,/* QOS MaxTRESRunMinsPerAccount
* exceeded (billing) */
WAIT_QOS_MAX_CPU_RUN_MINS_PER_ACCT,/* QOS MaxTRESRunMinsPerAccount
* exceeded (CPU) */
WAIT_QOS_MAX_ENERGY_RUN_MINS_PER_ACCT,/* QOS MaxTRESRunMinsPerAccount
* exceeded (Energy) */
WAIT_QOS_MAX_GRES_RUN_MINS_PER_ACCT,/* QOS MaxTRESRunMinsPerAccount
* exceeded (GRES) */
WAIT_QOS_MAX_NODE_RUN_MINS_PER_ACCT,/* QOS MaxTRESRunMinsPerAccount
* exceeded (Node) */
WAIT_QOS_MAX_LIC_RUN_MINS_PER_ACCT,/* QOS MaxTRESRunMinsPerAccount
* exceeded (license) */
WAIT_QOS_MAX_MEM_RUN_MINS_PER_ACCT,/* QOS MaxTRESRunMinsPerAccount
* exceeded (Memory) */
WAIT_QOS_MAX_UNK_RUN_MINS_PER_ACCT,/* QOS MaxTRESRunMinsPerAccount
* exceeded (Unknown) */
WAIT_QOS_MAX_BB_RUN_MINS_PER_USER,/* QOS MaxTRESRunMinsPerUser exceeded
* (burst buffer) */
WAIT_QOS_MAX_BILLING_RUN_MINS_PER_USER,/* QOS MaxTRESRunMinsPerUser
* exceeded (billing) */
WAIT_QOS_MAX_CPU_RUN_MINS_PER_USER,/* QOS MaxTRESRunMinsPerUser exceeded
* (CPU) */
WAIT_QOS_MAX_ENERGY_RUN_MINS_PER_USER,/* QOS MaxTRESRunMinsPerUser
* exceeded (Energy) */
WAIT_QOS_MAX_GRES_RUN_MINS_PER_USER,/* QOS MaxTRESRunMinsPerUser
* exceeded (GRES) */
WAIT_QOS_MAX_NODE_RUN_MINS_PER_USER,/* QOS MaxTRESRunMinsPerUser
* exceeded (Node) */
WAIT_QOS_MAX_LIC_RUN_MINS_PER_USER,/* QOS MaxTRESRunMinsPerUser exceeded
* (license) */
WAIT_QOS_MAX_MEM_RUN_MINS_PER_USER,/* QOS MaxTRESRunMinsPerUser exceeded
* (Memory) */
WAIT_QOS_MAX_UNK_RUN_MINS_PER_USER,/* QOS MaxTRESRunMinsPerUser exceeded
* (Unknown) */
WAIT_MAX_POWERED_NODES, /* max_powered_nodes reached */
WAIT_MPI_PORTS_BUSY, /* MPI resv_ports busy */
REASON_END, /* end of table */
};
enum job_acct_types {
JOB_START,
JOB_STEP,
JOB_SUSPEND,
JOB_TERMINATED
};
/* Partition state flags */
#define PARTITION_SUBMIT 0x01 /* Allow job submission to partition */
#define PARTITION_SCHED 0x02 /* Allow job startup from partition */
/* Actual partition states based upon state flags */
#define PARTITION_DOWN (PARTITION_SUBMIT)
#define PARTITION_UP (PARTITION_SUBMIT | PARTITION_SCHED)
#define PARTITION_DRAIN (PARTITION_SCHED)
#define PARTITION_INACTIVE 0x00
/* Partition enforce flags for jobs */
#define PARTITION_ENFORCE_NONE 0
#define PARTITION_ENFORCE_ALL 1 /* job limit must be valid for ALL
* partitions */
#define PARTITION_ENFORCE_ANY 2 /* job limit must be valid for ANY
* partition */
/*
* Auth plugin (id) used for communication.
* Update auth_plugin_types in slurm_auth.c if changed.
*/
enum auth_plugin_type {
AUTH_PLUGIN_NONE = 100,
AUTH_PLUGIN_MUNGE = 101,
AUTH_PLUGIN_JWT = 102,
AUTH_PLUGIN_SLURM = 103,
};
/*
* Hash plugin (id) used for communication.
*/
enum hash_plugin_type {
HASH_PLUGIN_DEFAULT = 0,
HASH_PLUGIN_NONE,
HASH_PLUGIN_K12,
HASH_PLUGIN_SHA256,
HASH_PLUGIN_SHA3,
HASH_PLUGIN_CNT,
};
/*
* TLS plugin (id) used for communication.
*/
enum tls_plugin_type {
TLS_PLUGIN_NONE = 100,
TLS_PLUGIN_S2N = 101,
};
/* Select plugin (id) in use by cluster */
enum select_plugin_type {
/* 100 unused (originally for BlueGene) */
/*
* 101 cons_res was removed in 23.11. However, this is needed for
* systems that are upgrading from an older version and were using
* cons_res. This can be removed two versions after 23.11.
*/
SELECT_PLUGIN_CONS_RES = 101,
SELECT_PLUGIN_LINEAR = 102, /* Linear on a normal system */
/* 103 unused (originally used for BGQ) */
/* 104 unused (originally used for Cray/ALPS with select/linear) */
/* 105 unused (originally used for Cray/ALPS with select/cons_res) */
/* 106 unused (originally used for Serial) */
/* 107 unused (originally used for Cray Aries with select/linear) */
/* 108 unused (originally used for Cray Aries with select/cons_res) */
SELECT_PLUGIN_CONS_TRES = 109, /* Cons TRES on a normal system */
/* 110 unused (originally used for Cray Aries with select/cons_tres) */
};
/* switch plugin (id) in use by cluster */
enum switch_plugin_type {
SWITCH_PLUGIN_NONE = 100, /* NONE */
SWITCH_PLUGIN_GENERIC = 101, /* Generic */
SWITCH_PLUGIN_CRAY = 102, /* Cray */
/* 103 unused (originally used for NRT) */
SWITCH_PLUGIN_SLINGSHOT = 104, /* HPE Slingshot */
SWITCH_PLUGIN_NVIDIA_IMEX = 105,
};
enum select_jobdata_type {
SELECT_JOBDATA_NETWORK = 23, /* data-> char * network info */
};
enum select_nodedata_type {
SELECT_NODEDATA_SUBCNT = 2, /* data-> uint16_t */
SELECT_NODEDATA_PTR = 5, /* data-> select_nodeinfo_t *nodeinfo */
SELECT_NODEDATA_MEM_ALLOC = 8, /* data-> uint32_t */
SELECT_NODEDATA_TRES_ALLOC_FMT_STR = 9, /* data-> char *,
* free with xfree */
SELECT_NODEDATA_TRES_ALLOC_WEIGHTED = 10, /* data-> double */
};
enum select_print_mode {
SELECT_PRINT_HEAD, /* Print just the header */
SELECT_PRINT_DATA, /* Print just the data */
SELECT_PRINT_MIXED, /* Print "field=value" */
SELECT_PRINT_MIXED_SHORT,/* Print less "field=value" */
SELECT_PRINT_BG_ID, /* Print just the BG_ID */
SELECT_PRINT_NODES, /* Print the nodelist */
SELECT_PRINT_CONNECTION,/* Print just the CONNECTION type */
SELECT_PRINT_ROTATE, /* Print just the ROTATE */
SELECT_PRINT_GEOMETRY, /* Print just the GEO */
SELECT_PRINT_START, /* Print just the START location */
SELECT_PRINT_BLRTS_IMAGE,/* Print just the BLRTS IMAGE */
SELECT_PRINT_LINUX_IMAGE,/* Print just the LINUX IMAGE */
SELECT_PRINT_MLOADER_IMAGE,/* Print just the MLOADER IMAGE */
SELECT_PRINT_RAMDISK_IMAGE,/* Print just the RAMDISK IMAGE */
SELECT_PRINT_REBOOT, /* Print just the REBOOT */
SELECT_PRINT_RESV_ID, /* Print just Cray/BASIL reservation ID */
SELECT_PRINT_START_LOC /* Print just the start location */
};
enum select_node_cnt {
SELECT_GET_NODE_SCALING, /* Give scaling factor for node count */
SELECT_GET_NODE_CPU_CNT, /* Give how many cpus are on a node */
SELECT_GET_MP_CPU_CNT, /* Give how many cpus are on a
* base partition */
SELECT_APPLY_NODE_MIN_OFFSET, /* Apply min offset to variable */
SELECT_APPLY_NODE_MAX_OFFSET, /* Apply max offset to variable */
SELECT_SET_NODE_CNT, /* Set altered node cnt */
SELECT_SET_MP_CNT /* Given a node cnt return the
* base partition count */
};
enum acct_gather_profile_info {
ACCT_GATHER_PROFILE_DIR, /* Give directory profiling is stored */
ACCT_GATHER_PROFILE_DEFAULT, /* What is being collected for
* profiling by default */
ACCT_GATHER_PROFILE_RUNNING /* What is actually be collected
* wither it be user or
* default. (Only works in the slurmstepd)
*/
};
#define ACCT_GATHER_PROFILE_NOT_SET 0x00000000
#define ACCT_GATHER_PROFILE_NONE SLURM_BIT(0)
#define ACCT_GATHER_PROFILE_ENERGY SLURM_BIT(1)
#define ACCT_GATHER_PROFILE_TASK SLURM_BIT(2)
#define ACCT_GATHER_PROFILE_LUSTRE SLURM_BIT(3)
#define ACCT_GATHER_PROFILE_NETWORK SLURM_BIT(4)
#define ACCT_GATHER_PROFILE_ALL 0xffffffff
/* jobacct data types */
enum jobacct_data_type {
JOBACCT_DATA_TOTAL, /* data-> jobacctinfo_t * */
JOBACCT_DATA_PIPE, /* data-> file descriptor */
JOBACCT_DATA_RUSAGE, /* data-> rusage set user_cpu_sec,
* user_cpu_usec, sys_cpu_sec, sys_cpu_usec */
JOBACCT_DATA_TOT_VSIZE = 5, /* data-> uint64_t vsize */
JOBACCT_DATA_TOT_RSS = 8, /* data-> uint64_t psize */
};
enum acct_energy_type {
ENERGY_DATA_JOULES_TASK,
ENERGY_DATA_STRUCT,
ENERGY_DATA_RECONFIG,
ENERGY_DATA_PROFILE,
ENERGY_DATA_LAST_POLL,
ENERGY_DATA_SENSOR_CNT,
ENERGY_DATA_NODE_ENERGY,
ENERGY_DATA_NODE_ENERGY_UP,
ENERGY_DATA_STEP_PTR
};
typedef enum {
UPDATE_SET, /* Set to specified value */
UPDATE_ADD, /* Append to existing value (+=)*/
UPDATE_REMOVE, /* Remove from existing vale (-=) */
} update_mode_t;
/*
* Task distribution states/methods
*
* Symbol format is SLURM_DIST_<node>_<socket>_<core>
*
* <node> = Method for distributing tasks to nodes.
* This determines the order in which task ids are
* distributed to the nodes selected for the job/step.
* <socket> = Method for distributing allocated lllps across sockets.
* This determines the order in which allocated lllps are
* distributed across sockets for binding to tasks.
* <core> = Method for distributing allocated lllps across cores.
* This determines the order in which allocated lllps are
* distributed across cores for binding to tasks.
*
* Note that the socket and core distributions apply only to task affinity.
*/
typedef enum task_dist_states {
/* NOTE: start SLURM_DIST_CYCLIC at 1 for HP MPI */
SLURM_DIST_CYCLIC = 0x0001,
SLURM_DIST_BLOCK = 0x0002,
SLURM_DIST_ARBITRARY = 0x0003,
SLURM_DIST_PLANE = 0x0004,
SLURM_DIST_CYCLIC_CYCLIC = 0x0011,
SLURM_DIST_CYCLIC_BLOCK = 0x0021,
SLURM_DIST_CYCLIC_CFULL = 0x0031,
SLURM_DIST_BLOCK_CYCLIC = 0x0012,
SLURM_DIST_BLOCK_BLOCK = 0x0022,
SLURM_DIST_BLOCK_CFULL = 0x0032,
SLURM_DIST_CYCLIC_CYCLIC_CYCLIC = 0x0111,
SLURM_DIST_CYCLIC_CYCLIC_BLOCK = 0x0211,
SLURM_DIST_CYCLIC_CYCLIC_CFULL = 0x0311,
SLURM_DIST_CYCLIC_BLOCK_CYCLIC = 0x0121,
SLURM_DIST_CYCLIC_BLOCK_BLOCK = 0x0221,
SLURM_DIST_CYCLIC_BLOCK_CFULL = 0x0321,
SLURM_DIST_CYCLIC_CFULL_CYCLIC = 0x0131,
SLURM_DIST_CYCLIC_CFULL_BLOCK = 0x0231,
SLURM_DIST_CYCLIC_CFULL_CFULL = 0x0331,
SLURM_DIST_BLOCK_CYCLIC_CYCLIC = 0x0112,
SLURM_DIST_BLOCK_CYCLIC_BLOCK = 0x0212,
SLURM_DIST_BLOCK_CYCLIC_CFULL = 0x0312,
SLURM_DIST_BLOCK_BLOCK_CYCLIC = 0x0122,
SLURM_DIST_BLOCK_BLOCK_BLOCK = 0x0222,
SLURM_DIST_BLOCK_BLOCK_CFULL = 0x0322,
SLURM_DIST_BLOCK_CFULL_CYCLIC = 0x0132,
SLURM_DIST_BLOCK_CFULL_BLOCK = 0x0232,
SLURM_DIST_BLOCK_CFULL_CFULL = 0x0332,
SLURM_DIST_NODECYCLIC = 0x0001,
SLURM_DIST_NODEBLOCK = 0x0002,
SLURM_DIST_SOCKCYCLIC = 0x0010,
SLURM_DIST_SOCKBLOCK = 0x0020,
SLURM_DIST_SOCKCFULL = 0x0030,
SLURM_DIST_CORECYCLIC = 0x0100,
SLURM_DIST_COREBLOCK = 0x0200,
SLURM_DIST_CORECFULL = 0x0300,
/* Unused = 0x1000, */
SLURM_DIST_UNKNOWN = 0x2000
} task_dist_states_t;
#define SLURM_DIST_STATE_BASE 0x00FFFF
#define SLURM_DIST_STATE_FLAGS 0xFF0000
#define SLURM_DIST_PACK_NODES 0x800000
#define SLURM_DIST_NO_PACK_NODES 0x400000
#define SLURM_DIST_NODEMASK 0xF00F
#define SLURM_DIST_SOCKMASK 0xF0F0
#define SLURM_DIST_COREMASK 0xFF00
#define SLURM_DIST_NODESOCKMASK 0xF0FF
/* Open stdout/err file mode, 0 for system default (JobFileAppend) */
#define OPEN_MODE_APPEND 1
#define OPEN_MODE_TRUNCATE 2
typedef enum cpu_bind_type { /* cpu binding type from --cpu-bind=... */
/* verbose can be set with any other flag */
CPU_BIND_VERBOSE = 0x0001, /* =v, */
/* the following auto-binding flags are mutually exclusive */
CPU_BIND_TO_THREADS = 0x0002, /* =threads */
CPU_BIND_TO_CORES = 0x0004, /* =cores */
CPU_BIND_TO_SOCKETS = 0x0008, /* =sockets */
CPU_BIND_TO_LDOMS = 0x0010, /* locality domains */
/* the following manual binding flags are mutually exclusive */
/* CPU_BIND_NONE needs to be the lowest value among manual bindings */
CPU_BIND_NONE = 0x0020, /* =no */
/* CPU_BIND_RANK = 0x0040 was removed in 24.11 */
CPU_BIND_MAP = 0x0080, /* =map_cpu:<list of CPU IDs> */
CPU_BIND_MASK = 0x0100, /* =mask_cpu:<list of CPU masks> */
CPU_BIND_LDRANK = 0x0200, /* =locality domain rank */
CPU_BIND_LDMAP = 0x0400, /* =map_ldom:<list of locality domains> */
CPU_BIND_LDMASK = 0x0800, /* =mask_ldom:<list of ldom masks> */
/* the following is used primarily for the
--hint=nomultithread when -mblock:block is requested. */
CPU_BIND_ONE_THREAD_PER_CORE = 0x2000,/* Only bind to one
* thread of a core */
/* default binding if auto binding doesn't match. */
CPU_AUTO_BIND_TO_THREADS = 0x04000,
CPU_AUTO_BIND_TO_CORES = 0x10000,
CPU_AUTO_BIND_TO_SOCKETS = 0x20000,
/* the following is used only as a flag for expressing
* the contents of TaskPluginParams */
SLURMD_OFF_SPEC = 0x40000,
CPU_BIND_OFF = 0x80000, /* Disable binding */
OOM_KILL_STEP = 0x100000
} cpu_bind_type_t;
#define CPU_BIND_T_TO_MASK 0x001e
#define CPU_BIND_T_AUTO_TO_MASK 0x34000
#define CPU_BIND_T_MASK 0x0fe0
#define CPU_BIND_T_TASK_PARAMS_MASK (SLURMD_OFF_SPEC | CPU_BIND_OFF)
/* Flag to indicate that cpu_freq is a range: low,medium,high,high-1
* instead of an integer value in kilohertz */
#define CPU_FREQ_RANGE_FLAG 0x80000000
#define CPU_FREQ_LOW 0x80000001
#define CPU_FREQ_MEDIUM 0x80000002
#define CPU_FREQ_HIGH 0x80000003
#define CPU_FREQ_HIGHM1 0x80000004
#define CPU_FREQ_CONSERVATIVE 0x88000000
#define CPU_FREQ_ONDEMAND 0x84000000
#define CPU_FREQ_PERFORMANCE 0x82000000
#define CPU_FREQ_POWERSAVE 0x81000000
#define CPU_FREQ_USERSPACE 0x80800000
#define CPU_FREQ_SCHEDUTIL 0x80400000
#define CPU_FREQ_GOV_MASK 0x8ff00000
typedef enum mem_bind_type { /* memory binding type from --mem-bind=... */
/* verbose can be set with any other flag */
MEM_BIND_VERBOSE= 0x01, /* =v, */
/* the following five manual binding flags are mutually exclusive */
/* MEM_BIND_NONE needs to be the first in this sub-list */
MEM_BIND_NONE = 0x02, /* =no */
MEM_BIND_RANK = 0x04, /* =rank */
MEM_BIND_MAP = 0x08, /* =map_mem:<list of NUMA IDs> */
MEM_BIND_MASK = 0x10, /* =mask_mem:<list of NUMA masks> */
MEM_BIND_LOCAL = 0x20, /* =local */
/* sort and prefer can be set with any other flags */
MEM_BIND_SORT = 0x40, /* =sort */
MEM_BIND_PREFER = 0x80 /* =prefer */
} mem_bind_type_t;
#define MEM_BIND_TYPE_MASK 0x3e
#define MEM_BIND_TYPE_FLAGS_MASK 0xc0
typedef enum accel_bind_type { /* accelerator binding from --accel_bind= */
ACCEL_BIND_VERBOSE = 0x01, /* 'v' verbose */
ACCEL_BIND_CLOSEST_GPU = 0x02, /* 'g' Use closest GPU to the CPU */
ACCEL_BIND_CLOSEST_NIC = 0x08 /* 'n' Use closest NIC to CPU */
} accel_bind_type_t;
/* The last entry in node_states must be STATE_END, keep in sync with
* node_state_string. values may be ORed with NODE_STATE_FLAGS below.
* Node states typically alternate between NODE_STATE_IDLE and
* NODE_STATE_ALLOCATED. The NODE_STATE_COMPLETING flag will be set
* when jobs are in the process of terminating. */
enum node_states {
NODE_STATE_UNKNOWN, /* node's initial state, unknown */
NODE_STATE_DOWN, /* node in non-usable state */
NODE_STATE_IDLE, /* node idle and available for use */
NODE_STATE_ALLOCATED, /* node has been allocated to a job */
NODE_STATE_ERROR, /* UNUSED - node is in an error state */
NODE_STATE_MIXED, /* node has a mixed state */
NODE_STATE_FUTURE, /* node slot reserved for future use */
NODE_STATE_END /* last entry in table */
};
#define NODE_STATE_BASE 0x0000000f
#define NODE_STATE_FLAGS 0xfffffff0
/* SLURM_BIT(0-3) taken for base states */
/* was NODE_STATE_NET SLURM_BIT(4), removed v24.05 */
#define NODE_STATE_RES SLURM_BIT(5) /* If a node is in a
* reservation (used primarily
* to note a node isn't idle
* for non-reservation jobs) */
#define NODE_STATE_UNDRAIN SLURM_BIT(6) /* Clear DRAIN flag for a node */
#define NODE_STATE_CLOUD SLURM_BIT(7) /* node comes from cloud */
#define NODE_RESUME SLURM_BIT(8) /* Restore a DRAINED, DRAINING, DOWN
* or FAILING node to service (e.g.
* IDLE or ALLOCATED). Used in
* slurm_update_node() request */
#define NODE_STATE_DRAIN SLURM_BIT(9) /* do not allocated new work */
#define NODE_STATE_COMPLETING SLURM_BIT(10) /* node is completing allocated
* job */
#define NODE_STATE_NO_RESPOND SLURM_BIT(11) /* node is not responding */
#define NODE_STATE_POWERED_DOWN SLURM_BIT(12) /* node is powered down */
#define NODE_STATE_FAIL SLURM_BIT(13) /* node is failing, do not allocate
* new work */
#define NODE_STATE_POWERING_UP SLURM_BIT(14) /* node is powering up */
#define NODE_STATE_MAINT SLURM_BIT(15) /* node in maintenance
* reservation */
#define NODE_STATE_REBOOT_REQUESTED SLURM_BIT(16) /* node reboot requested */
#define NODE_STATE_REBOOT_CANCEL SLURM_BIT(17) /* cancel pending reboot */
#define NODE_STATE_POWERING_DOWN SLURM_BIT(18) /* node is powering down */
#define NODE_STATE_DYNAMIC_FUTURE SLURM_BIT(19) /* dynamic future node */
#define NODE_STATE_REBOOT_ISSUED SLURM_BIT(20) /* node reboot passed to agent */
#define NODE_STATE_PLANNED SLURM_BIT(21) /* node scheduled for a job in the
* future */
#define NODE_STATE_INVALID_REG SLURM_BIT(22) /* invalid registration, don't
* ping */
#define NODE_STATE_POWER_DOWN SLURM_BIT(23) /* manual node power down */
#define NODE_STATE_POWER_UP SLURM_BIT(24) /* manual node power up */
#define NODE_STATE_POWER_DRAIN SLURM_BIT(25) /* signal power down asap */
#define NODE_STATE_DYNAMIC_NORM SLURM_BIT(26) /* dynamic norm node */
#define NODE_STATE_BLOCKED SLURM_BIT(27) /* node blocked by exclusive topo */
/* Used as show_flags for slurm_get_ and slurm_load_ function calls.
* Values can be ORed */
#define SHOW_ALL SLURM_BIT(0) /* Show info for "hidden" partitions */
#define SHOW_DETAIL SLURM_BIT(1) /* Show detailed resource information */
/* SLURM_BIT(2) empty */
#define SHOW_MIXED SLURM_BIT(3) /* Automatically set node MIXED state */
#define SHOW_LOCAL SLURM_BIT(4) /* Show only local information, even on
* federated cluster */
#define SHOW_SIBLING SLURM_BIT(5) /* Show sibling jobs on a federated cluster */
#define SHOW_FEDERATION SLURM_BIT(6) /* Show federated state information.
* Shows local info if not in federation */
#define SHOW_FUTURE SLURM_BIT(7) /* Show future nodes */
/* CR_CPU, CR_SOCKET and CR_CORE are mutually exclusive
* CR_MEMORY may be added to any of the above values or used by itself
* CR_ONE_TASK_PER_CORE may also be added to any of the above values */
#define CR_CPU 0x0001 /* Resources are shared down to the level of
* logical processors which can be socket,
* core, or thread depending on the system. */
#define CR_SOCKET 0x0002 /* Resources are shared down to the socket
* level. Jobs will not be co-allocated
* within a socket. */
#define CR_CORE 0x0004 /* Resources are shared down to the core level.
* Jobs will not be co-allocated within a
* core. */
#define CR_BOARD 0x0008 /* Resources are shared down to the board
* level. Jobs will not be co-allocated
* within a board. */
#define CR_MEMORY 0x0010 /* Memory as consumable resources. Memory is
* not over-committed when selected as a CR. */
/* was CR_OTHER_CONS_RES 0x0020, removed v23.11 */
#define ENFORCE_BINDING_GRES 0x0040
#define ONE_TASK_PER_SHARING_GRES 0x0080
/* By default, schedule only one task per core.
* Without this option, tasks would be allocated threads. */
#define CR_ONE_TASK_PER_CORE 0x0100
/* Pack tasks tightly onto allocated nodes rather than distributing them evenly
* across available nodes */
#define CR_PACK_NODES 0x0200
#define LL_SHARED_GRES 0x0400 /* Prefer least-loaded device for shared GRES */
/* was CR_OTHER_CONS_TRES 0x0800, removed v24.05 */
/* By default, distribute cores using a block approach inside the nodes */
#define CR_CORE_DEFAULT_DIST_BLOCK 0x1000
#define CR_LLN 0x4000 /* Select nodes by "least loaded." */
#define MULTIPLE_SHARING_GRES_PJ 0x8000 /* Allow multiple sharing gres per job */
/*
* This is used internally to know whether the job was started with
* cons_tres or linear. It is not a configuration option.
*/
#define CR_LINEAR 0x8000
#define MEM_PER_CPU 0x8000000000000000
#define SHARED_FORCE 0x8000
#define PRIVATE_DATA_JOBS SLURM_BIT(0) /* job/step data is private */
#define PRIVATE_DATA_NODES SLURM_BIT(1) /* node data is private */
#define PRIVATE_DATA_PARTITIONS SLURM_BIT(2) /* partition data is private */
#define PRIVATE_DATA_USAGE SLURM_BIT(3) /* accounting usage data is
* private */
#define PRIVATE_DATA_USERS SLURM_BIT(4) /* accounting user data is
* private */
#define PRIVATE_DATA_ACCOUNTS SLURM_BIT(5) /* accounting account data is
* private */
#define PRIVATE_DATA_RESERVATIONS SLURM_BIT(6) /* reservation data is private */
/* SLURM_BIT(7) Available 2 versions after 23.02 */
#define PRIVATE_DATA_EVENTS SLURM_BIT(8) /* events are private */
#define PRIORITY_RESET_NONE 0x0000 /* never clear */
#define PRIORITY_RESET_NOW 0x0001 /* clear now (when slurmctld restarts) */
#define PRIORITY_RESET_DAILY 0x0002 /* clear daily at midnight */
#define PRIORITY_RESET_WEEKLY 0x0003 /* clear weekly at Sunday 00:00 */
#define PRIORITY_RESET_MONTHLY 0x0004 /* clear monthly on first at 00:00 */
#define PRIORITY_RESET_QUARTERLY 0x0005 /* clear quarterly on first at 00:00 */
#define PRIORITY_RESET_YEARLY 0x0006 /* clear yearly on first at 00:00 */
#define PROP_PRIO_OFF 0x0000 /* Do not propagage user nice value */
#define PROP_PRIO_ON 0x0001 /* Propagate user nice value */
#define PROP_PRIO_NICER 0x0002 /* Ensure that user tasks have a nice
* value that is higher than slurmd */
#define PRIORITY_FLAGS_ACCRUE_ALWAYS SLURM_BIT(0) /* Flag to always accrue
* age priority to pending
* jobs ignoring
* dependencies or holds
*/
#define PRIORITY_FLAGS_MAX_TRES SLURM_BIT(1) /* Calculate billed_tres
* as the MAX of TRES on a
* node rather than the
* sum or TRES. */
#define PRIORITY_FLAGS_SIZE_RELATIVE SLURM_BIT(2) /* Enable job size
* measurement relative to
* its time limit */
#define PRIORITY_FLAGS_DEPTH_OBLIVIOUS SLURM_BIT(3) /* Flag to use depth
* oblivious formula for
* computing hierarchical
* fairshare */
#define PRIORITY_FLAGS_CALCULATE_RUNNING SLURM_BIT(4) /* Calculate priorities
* for running jobs, not
* only the pending jobs.
*/
#define PRIORITY_FLAGS_FAIR_TREE SLURM_BIT(5) /* Prioritize by level in
* account hierarchy. */
#define PRIORITY_FLAGS_INCR_ONLY SLURM_BIT(6) /* Priority can only
* increase, never
* decrease in value */
#define PRIORITY_FLAGS_NO_NORMAL_ASSOC SLURM_BIT(7)
#define PRIORITY_FLAGS_NO_NORMAL_PART SLURM_BIT(8)
#define PRIORITY_FLAGS_NO_NORMAL_QOS SLURM_BIT(9)
#define PRIORITY_FLAGS_NO_NORMAL_TRES SLURM_BIT(10)
/*
* These bits are set in the bitflags field of job_desc_msg_t and
* job_flags in slurm_opt_t
*/
#define KILL_INV_DEP SLURM_BIT(0) /* Kill job on invalid dependency */
#define NO_KILL_INV_DEP SLURM_BIT(1) /* Don't kill job on invalid
* dependency */
#define HAS_STATE_DIR SLURM_BIT(2) /* Used by slurmctld to track
* state dir */
#define BACKFILL_TEST SLURM_BIT(3) /* Backfill test in progress */
#define GRES_ENFORCE_BIND SLURM_BIT(4) /* Enforce CPU/GRES binding */
#define TEST_NOW_ONLY SLURM_BIT(5) /* Test for immediately start only */
#define JOB_SEND_ENV SLURM_BIT(6) /* Send env to the dbd */
#define GRACE_PREEMPT SLURM_BIT(7) /* Job signalled and waiting for
Gracetime */
#define SPREAD_JOB SLURM_BIT(8) /* Spread job across max node count */
#define USE_MIN_NODES SLURM_BIT(9) /* Prefer minimum node count */
#define JOB_KILL_HURRY SLURM_BIT(10) /* Avoid burst buffer stage out */
#define TRES_STR_CALC SLURM_BIT(11) /* Avoid calculating TRES strings at
* the end of a job. */
#define SIB_JOB_FLUSH SLURM_BIT(12) /* Don't send complete to origin */
#define HET_JOB_FLAG SLURM_BIT(13) /* Heterogeneous job management flag */
#define JOB_NTASKS_SET SLURM_BIT(14) /* --ntasks explicitly set */
#define JOB_CPUS_SET SLURM_BIT(15) /* --cpus-per-tasks explicitly set */
#define BF_WHOLE_NODE_TEST SLURM_BIT(16) /* Backfill test in progress */
#define TOP_PRIO_TMP SLURM_BIT(17) /* Temporary flag for top priority job
* operation */
#define JOB_ACCRUE_OVER SLURM_BIT(18) /* We have cleared the accrual count of
* a job. */
#define GRES_DISABLE_BIND SLURM_BIT(19) /* Disable CPU/GRES binding */
#define JOB_WAS_RUNNING SLURM_BIT(20) /* Job was running */
#define RESET_ACCRUE_TIME SLURM_BIT(21) /* Reset the job's accrue time */
#define CRON_JOB SLURM_BIT(22) /* Job submitted through scrontab */
#define JOB_MEM_SET SLURM_BIT(23) /* Memory limit explicitly set by job */
/* SLURM_BIT(24) Removed in 23.11, free in 25.05 */
#define USE_DEFAULT_ACCT SLURM_BIT(25) /* Job submitted to default account */
#define USE_DEFAULT_PART SLURM_BIT(26) /* Job submitted to default
* partition */
#define USE_DEFAULT_QOS SLURM_BIT(27) /* Job submitted with default QOS */
#define USE_DEFAULT_WCKEY SLURM_BIT(28) /* Job submitted with default WCKEY */
#define JOB_DEPENDENT SLURM_BIT(29) /* Job dependent or invalid depend */
#define JOB_MAGNETIC SLURM_BIT(30) /* Job attempting to run in a
* magnetic reservation */
#define JOB_PART_ASSIGNED SLURM_BIT(31) /* Job didn't request a partition */
#define BACKFILL_SCHED SLURM_BIT(32) /* Job was considered in last
* backfill attempt if not set the
* normal scheduler set
* last_eval_time */
#define BACKFILL_LAST SLURM_BIT(33) /* Job was considered in last
* schedule attempt */
#define TASKS_CHANGED SLURM_BIT(34) /* Reset licenses per job */
#define JOB_SEND_SCRIPT SLURM_BIT(35) /* Send script to the dbd */
#define RESET_LIC_TASK SLURM_BIT(36) /* Reset licenses per task */
#define RESET_LIC_JOB SLURM_BIT(37) /* Reset licenses per job */
#define GRES_ONE_TASK_PER_SHARING SLURM_BIT(38) /* Don't let tasks in the same
* job share the same gres */
#define GRES_MULT_TASKS_PER_SHARING SLURM_BIT(39)/* Negate
* GRES_ONE_TASK_PER_SHARING */
#define GRES_ALLOW_TASK_SHARING SLURM_BIT(40) /* Allow tasks to share gres */
#define STEPMGR_ENABLED SLURM_BIT(41) /* enable(d) stepmgr or not */
/* These bits are set in the x11 field of job_desc_msg_t */
#define X11_FORWARD_ALL 0x0001 /* all nodes should setup forward */
#define X11_FORWARD_BATCH 0x0002 /* only the batch node */
#define X11_FORWARD_FIRST 0x0004 /* only the first node */
#define X11_FORWARD_LAST 0x0008 /* only the last node */
typedef enum {
SSF_NONE = 0, /* No flags set */
SSF_EXCLUSIVE = 1 << 0, /* CPUs not shared with other steps */
SSF_NO_KILL = 1 << 1, /* Don't kill step on node failure */
SSF_OVERCOMMIT = 1 << 2, /* Allow the step allocation of more tasks
* to a node than available processors. */
SSF_WHOLE = 1 << 3, /* Use whole nodes in step allocation */
SSF_INTERACTIVE = 1 << 4, /* Request interactive step allocation */
SSF_MEM_ZERO = 1 << 5, /* Requested --mem=0; use all memory but do not
* count against the job's memory allocation */
SSF_OVERLAP_FORCE = 1 << 6, /* Force this to overlap with all other
* steps; resources allocated to this step
* are not decremented from the job's
* allocation */
SSF_NO_SIG_FAIL = 1 << 7, /* Don't fail step due to signal */
SSF_EXT_LAUNCHER = 1 << 8, /* Request is for an external launcher */
SSF_GRES_ALLOW_TASK_SHARING = 1 << 9,
} step_spec_flags_t;
enum topology_plugin_type {
TOPOLOGY_PLUGIN_DEFAULT = 100,
TOPOLOGY_PLUGIN_3DTORUS = 101,
TOPOLOGY_PLUGIN_TREE = 102,
TOPOLOGY_PLUGIN_BLOCK = 103,
};
/*****************************************************************************\
* SLURM LIBRARY INITIALIZATION FUNCTIONS
\*****************************************************************************/
/*
* MUST be called before any other Slurm library API calls.
*
* conf should be a fully qualified path to a slurm.conf configuration file,
* or more commonly NULL to allow libslurm to automatically locate its own
* configuration.
*/
extern void slurm_init(const char *conf);
/*
* Call at process termination to cleanup internal configuration structures.
*
* Strongly recommended if valgrind or similar tools will be used to check
* your application for memory leaks.
*/
extern void slurm_fini(void);
/*****************************************************************************\
* SLURM HOSTLIST FUNCTIONS
\*****************************************************************************/
/* The hostlist opaque data type
*
* A hostlist is a list of hostnames optimized for a prefixXXXX style
* naming convention, where XXXX is a decimal, numeric suffix.
*/
#ifndef __hostlist_t_defined
# define __hostlist_t_defined
typedef struct hostlist hostlist_t;
#endif
/*
* slurm_hostlist_create():
*
* Create a new hostlist from a string representation.
*
* The string representation (str) may contain one or more hostnames or
* bracketed hostlists separated by either `,' or whitespace. A bracketed
* hostlist is denoted by a common prefix followed by a list of numeric
* ranges contained within brackets: e.g. "tux[0-5,12,20-25]"
*
* To support systems with 3-D topography, a rectangular prism may
* be described using two three digit numbers separated by "x": e.g.
* "bgl[123x456]". This selects all nodes between 1 and 4 inclusive
* in the first dimension, between 2 and 5 in the second, and between
* 3 and 6 in the third dimension for a total of 4*4*4=64 nodes
*
* Note: if this module is compiled with WANT_RECKLESS_HOSTRANGE_EXPANSION
* defined, a much more loose interpretation of host ranges is used.
* Reckless hostrange expansion allows all of the following (in addition to
* bracketed hostlists):
*
* o tux0-5,tux12,tux20-25
* o tux0-tux5,tux12,tux20-tux25
* o tux0-5,12,20-25
*
* If str is NULL, and empty hostlist is created and returned.
*
* If the create fails, hostlist_create() returns NULL.
*
* The returned hostlist must be freed with hostlist_destroy()
*
*/
extern hostlist_t *slurm_hostlist_create(const char *hostlist);
/* slurm_hostlist_count():
*
* Return the number of hosts in hostlist hl.
*/
extern int slurm_hostlist_count(hostlist_t *hl);
/*
* slurm_hostlist_destroy():
*
* Destroy a hostlist object. Frees all memory allocated to the hostlist.
*/
extern void slurm_hostlist_destroy(hostlist_t *hl);
/* slurm_hostlist_find():
*
* Searches hostlist hl for the first host matching hostname
* and returns position in list if found.
*
* Returns -1 if host is not found.
*/
extern int slurm_hostlist_find(hostlist_t *hl, const char *hostname);
/* slurm_hostlist_push():
*
* push a string representation of hostnames onto a hostlist.
*
* The hosts argument may take the same form as in slurm_hostlist_create()
*
* Returns the number of hostnames inserted into the list,
* or 0 on failure.
*/
extern int slurm_hostlist_push(hostlist_t *hl, const char *hosts);
/* slurm_hostlist_push_host():
*
* Push a single host onto the hostlist hl.
* This function is more efficient than slurm_hostlist_push() for a single
* hostname, since the argument does not need to be checked for ranges.
*
* return value is 1 for success, 0 for failure.
*/
extern int slurm_hostlist_push_host(hostlist_t *hl, const char *host);
/* slurm_hostlist_ranged_string():
*
* Write the string representation of the hostlist hl into buf,
* writing at most n chars. Returns the number of bytes written,
* or -1 if truncation occurred.
*
* The result will be NULL terminated.
*
* slurm_hostlist_ranged_string() will write a bracketed hostlist representation
* where possible.
*/
extern ssize_t slurm_hostlist_ranged_string(hostlist_t *hl, size_t n, char *buf);
/* hostlist_ranged_string_xmalloc():
*
* Wrapper of hostlist_ranged_string(), with result buffer dynamically
* allocated using xmalloc().
* The result will be NULL on failure (out of memory).
*
* Caller should free the result string using xfree().
*/
extern char *slurm_hostlist_ranged_string_xmalloc(hostlist_t *hl);
/*
* slurm_hostlist_shift():
*
* Returns the string representation of the first host in the hostlist
* or NULL if the hostlist is empty or there was an error allocating memory.
* The host is removed from the hostlist.
*
* Note: Caller is responsible for freeing the returned memory.
*/
extern char *slurm_hostlist_shift(hostlist_t *hl);
/* slurm_hostlist_uniq():
*
* Sort the hostlist hl and remove duplicate entries.
*
*/
extern void slurm_hostlist_uniq(hostlist_t *hl);
/*****************************************************************************\
* SLURM LIST FUNCTIONS
\*****************************************************************************/
#ifndef __list_datatypes_defined
# define __list_datatypes_defined
typedef struct xlist list_t;
/*
* List opaque data type.
*/
typedef struct listIterator list_itr_t;
/*
* List Iterator opaque data type.
*/
typedef void (*ListDelF) (void *x);
/*
* Function prototype to deallocate data stored in a list.
* This function is responsible for freeing all memory associated
* with an item, including all subordinate items (if applicable).
*/
typedef int (*ListCmpF) (void *x, void *y);
/*
* Function prototype for comparing two items in a list.
* Returns less-than-zero if (x<y), zero if (x==y), and
* greather-than-zero if (x>y).
*/
typedef int (*ListFindF) (void *x, void *key);
/*
* Function prototype for matching items in a list.
* Returns non-zero if (x==key); o/w returns zero.
*/
typedef int (*ListForF) (void *x, void *arg);
/*
* Function prototype for operating on each item in a list.
* Returns less-than-zero on error.
*/
#endif
/* slurm_list_append():
*
* Inserts data [x] at the end of list [l].
*/
extern void slurm_list_append(list_t *l, void *x);
/* slurm_list_count():
*
* Returns the number of items in list [l].
*/
extern int slurm_list_count(list_t *l);
/* slurm_list_create():
*
* Creates and returns a new empty list.
* The deletion function [f] is used to deallocate memory used by items
* in the list; if this is NULL, memory associated with these items
* will not be freed when the list is destroyed.
* Note: Abandoning a list without calling slurm_list_destroy() will result
* in a memory leak.
*/
extern list_t *slurm_list_create(ListDelF f);
/* slurm_list_destroy():
*
* Destroys list [l], freeing memory used for list iterators and the
* list itself; if a deletion function was specified when the list
* was created, it will be called for each item in the list.
*/
extern void slurm_list_destroy(list_t *l);
/* slurm_list_find():
*
* Traverses the list from the point of the list iterator [i]
* using [f] to match each item with [key].
* Returns a ptr to the next item for which the function [f]
* returns non-zero, or NULL once the end of the list is reached.
* Example: i=slurm_list_iterator_reset(i);
* while ((x=slurm_list_find(i,f,k))) {...}
*/
extern void *slurm_list_find(list_itr_t *i, ListFindF f, void *key);
/* slurm_list_is_empty():
*
* Returns non-zero if list [l] is empty; o/w returns zero.
*/
extern int slurm_list_is_empty(list_t *l);
/*
* Creates and returns a list iterator for non-destructively traversing
* list [l].
*/
extern list_itr_t *slurm_list_iterator_create(list_t *l);
/* slurm_list_iterator_reset():
*
* Resets the list iterator [i] to start traversal at the beginning
* of the list.
*/
extern void slurm_list_iterator_reset(list_itr_t *i);
/*
* Destroys the list iterator [i]; list iterators not explicitly destroyed
* in this manner will be destroyed when the list is deallocated via
* list_destroy().
*/
extern void slurm_list_iterator_destroy(list_itr_t *i);
/* slurm_list_next():
*
* Returns a ptr to the next item's data,
* or NULL once the end of the list is reached.
* Example: i=slurm_list_iterator_create(i);
* while ((x=slurm_list_next(i))) {...}
*/
extern void *slurm_list_next(list_itr_t *i);
/* slurm_list_sort():
*
* Sorts list [l] into ascending order according to the function [f].
* Note: Sorting a list resets all iterators associated with the list.
* Note: The sort algorithm is stable.
*/
extern void slurm_list_sort(list_t *l, ListCmpF f);
/* slurm_list_pop():
*
* Pops the data item at the top of the stack [l].
* Returns the data's ptr, or NULL if the stack is empty.
*/
extern void *slurm_list_pop(list_t *l);
/*****************************************************************************\
* SLURM BITSTR FUNCTIONS
\*****************************************************************************/
#ifndef __bitstr_datatypes_defined
# define __bitstr_datatypes_defined
typedef int64_t bitstr_t;
typedef bitstr_t bitoff_t;
#endif
#define ALLOC_SID_ADMIN_HOLD 0x00000001 /* admin job hold */
#define ALLOC_SID_USER_HOLD 0x00000002 /* user job hold */
#define JOB_SHARED_NONE 0x0000
#define JOB_SHARED_OK 0x0001
#define JOB_SHARED_USER 0x0002
#define JOB_SHARED_MCS 0x0003
#define JOB_SHARED_TOPO 0x0004
/*****************************************************************************\
* PROTOCOL DATA STRUCTURE DEFINITIONS
\*****************************************************************************/
typedef struct dynamic_plugin_data {
void *data;
uint32_t plugin_id;
} dynamic_plugin_data_t;
typedef struct acct_gather_energy {
uint32_t ave_watts; /* average power consump of node, in watts */
uint64_t base_consumed_energy;
uint64_t consumed_energy; /* total energy consumed by node, in joules */
uint32_t current_watts; /* current power consump of node, in watts */
uint64_t last_adjustment; /* last applied adjustment, in joules.
* DO NOT PACK. */
uint64_t previous_consumed_energy;
time_t poll_time; /* When information was last retrieved */
time_t slurmd_start_time; /* Slurmd start time */
} acct_gather_energy_t;
typedef struct {
time_t expiration;
char *net_cred;
slurm_addr_t *node_addrs;
uint32_t node_cnt;
char *node_list;
} slurm_node_alias_addrs_t;
#define CORE_SPEC_THREAD 0x8000 /* If set, this is a thread count not core count */
/*
* Update:
* _copy_job_desc_to_job_record()
* slurm_free_job_desc_msg()
*/
typedef struct job_descriptor { /* For submit, allocate, and update requests */
char *account; /* charge to specified account */
char *acctg_freq; /* accounting polling intervals (seconds) */
char *admin_comment; /* administrator's arbitrary comment (update only) */
char *alloc_node; /* node making resource allocation request
* NOTE: Normally set by slurm_submit* or
* slurm_allocate* function */
uint16_t alloc_resp_port; /* port to send allocation confirmation to */
uint32_t alloc_sid; /* local sid making resource allocation request
* NOTE: Normally set by slurm_submit* or
* slurm_allocate* function
* NOTE: Also used for update flags, see
* ALLOC_SID_* flags */
uint32_t argc; /* number of arguments to the script */
char **argv; /* arguments to the script */
char *array_inx; /* job array index values */
bitstr_t *array_bitmap; /* NOTE: Set by slurmctld */
char *batch_features; /* features required for batch script's node */
time_t begin_time; /* delay initiation until this time */
uint64_t bitflags; /* bitflags */
char *burst_buffer; /* burst buffer specifications */
char *clusters; /* cluster names used for multi-cluster jobs */
char *cluster_features; /* required cluster feature specification,
* default NONE */
char *comment; /* arbitrary comment */
uint16_t contiguous; /* 1 if job requires contiguous nodes,
* 0 otherwise,default=0 */
char *container; /* OCI container bundle */
char *container_id; /* OCI container ID */
uint16_t core_spec; /* specialized core/thread count,
* see CORE_SPEC_THREAD */
char *cpu_bind; /* binding map for map/mask_cpu - This
* currently does not matter to the
* job allocation, setting this does
* not do anything for steps. */
uint16_t cpu_bind_type; /* see cpu_bind_type_t - This
* currently does not matter to the
* job allocation, setting this does
* not do anything for steps. */
uint32_t cpu_freq_min; /* Minimum cpu frequency */
uint32_t cpu_freq_max; /* Maximum cpu frequency */
uint32_t cpu_freq_gov; /* cpu frequency governor */
char *cpus_per_tres; /* semicolon delimited list of TRES=# values */
void *crontab_entry; /* really cron_entry_t */
time_t deadline; /* deadline */
uint32_t delay_boot; /* delay boot for desired node state */
char *dependency; /* synchronize job execution with other jobs */
time_t end_time; /* time by which job must complete, used for
* job update only now, possible deadline
* scheduling in the future */
char **environment; /* environment variables to set for job,
* name=value pairs, one per line */
slurm_hash_t env_hash; /* hash value of environment NO NOT PACK */
uint32_t env_size; /* element count in environment */
char *exc_nodes; /* comma separated list of nodes excluded
* from job's allocation, default NONE */
char *extra; /* Arbitrary string */
char *features; /* required feature specification,
* default NONE */
uint64_t fed_siblings_active; /* Bitmap of active fed sibling ids */
uint64_t fed_siblings_viable; /* Bitmap of viable fed sibling ids */
uint32_t group_id; /* group to assume, if run as root. */
uint32_t het_job_offset; /* HetJob component offset */
void *id; /* actually identity_t. DO NOT PACK */
uint16_t immediate; /* 1 if allocate to run or fail immediately,
* 0 if to be queued awaiting resources */
uint32_t job_id; /* job ID, default set by Slurm */
char * job_id_str; /* string representation of the jobid */
char *job_size_str;
uint16_t kill_on_node_fail; /* 1 if node failure to kill job,
* 0 otherwise,default=1 */
char *licenses; /* licenses required by the job */
char *licenses_tot; /* total licenses required by the job included
* from tres requests as well, NOT PACKED */
uint16_t mail_type; /* see MAIL_JOB_ definitions above */
char *mail_user; /* user to receive notification */
char *mcs_label; /* mcs_label if mcs plugin in use */
char *mem_bind; /* binding map for map/mask_cpu */
uint16_t mem_bind_type; /* see mem_bind_type_t */
char *mem_per_tres; /* semicolon delimited list of TRES=# values */
char *name; /* name of the job, default "" */
char *network; /* network use spec */
uint32_t nice; /* requested priority change,
* NICE_OFFSET == no change */
uint32_t num_tasks; /* number of tasks to be started,
* for batch only */
uint16_t oom_kill_step; /* 1 if kill whole step on task OOM,
* 0 do nothing on task OOM
* NO_VAL16 use taskpluginparam option */
uint8_t open_mode; /* out/err open mode truncate or append,
* see OPEN_MODE_* */
char *origin_cluster; /* cluster name that initiated the job. */
uint16_t other_port; /* port to send various notification msg to */
uint8_t overcommit; /* over subscribe resources, for batch only */
char *partition; /* name of requested partition,
* default in Slurm config */
uint16_t plane_size; /* plane size when task_dist =
SLURM_DIST_PLANE */
char *prefer; /* soft feature specification,
* default NONE */
uint32_t priority; /* relative priority of the job,
* explicitly set only for user root,
* 0 == held (don't initiate) */
uint32_t profile; /* Level of acct_gather_profile {all | none} */
char *qos; /* Quality of Service */
uint16_t reboot; /* force node reboot before startup */
char *resp_host; /* NOTE: Set by slurmctld */
uint16_t restart_cnt; /* count of job restarts */
char *req_nodes; /* comma separated list of required nodes
* default NONE */
uint16_t requeue; /* enable or disable job requeue option */
char *reservation; /* name of reservation to use */
uint16_t resv_port_cnt; /* count of MPI ports reserved per node */
char *script; /* the actual job script, default NONE */
void *script_buf; /* job script as mmap buf */
slurm_hash_t script_hash; /* hash value of script NO NOT PACK */
uint16_t shared; /* 2 if the job can only share nodes with other
* jobs owned by that user,
* 1 if job can share nodes with other jobs,
* 0 if job needs exclusive access to the node,
* or NO_VAL to accept the system default.
* SHARED_FORCE to eliminate user control. */
uint32_t site_factor; /* factor to consider in priority */
char **spank_job_env; /* environment variables for job prolog/epilog
* scripts as set by SPANK plugins */
uint32_t spank_job_env_size; /* element count in spank_env */
char *submit_line; /* The command issued with all it's options in a
* string */
uint32_t task_dist; /* see enum task_dist_state */
uint32_t time_limit; /* maximum run time in minutes, default is
* partition limit */
uint32_t time_min; /* minimum run time in minutes, default is
* time_limit */
char *tres_bind; /* Task to TRES binding directives */
char *tres_freq; /* TRES frequency directives */
char *tres_per_job; /* semicolon delimited list of TRES=# values */
char *tres_per_node; /* semicolon delimited list of TRES=# values */
char *tres_per_socket; /* semicolon delimited list of TRES=# values */
char *tres_per_task; /* semicolon delimited list of TRES=# values */
uint32_t user_id; /* set only if different from current UID,
* can only be explicitly set by user root */
uint16_t wait_all_nodes;/* 0 to start job immediately after allocation
* 1 to start job after all nodes booted
* or NO_VAL to use system default */
uint16_t warn_flags; /* flags related to job signals
* (eg. KILL_JOB_BATCH) */
uint16_t warn_signal; /* signal to send when approaching end time */
uint16_t warn_time; /* time before end to send signal (seconds) */
char *work_dir; /* pathname of working directory */
/* job constraints: */
uint16_t cpus_per_task; /* number of processors required for
* each task */
uint32_t min_cpus; /* minimum number of processors required,
* default=0 */
uint32_t max_cpus; /* maximum number of processors required,
* default=0 */
uint32_t min_nodes; /* minimum number of nodes required by job,
* default=0 */
uint32_t max_nodes; /* maximum number of nodes usable by job,
* default=0 */
uint16_t boards_per_node; /* boards per node required by job */
uint16_t sockets_per_board;/* sockets per board required by job */
uint16_t sockets_per_node;/* sockets per node required by job */
uint16_t cores_per_socket;/* cores per socket required by job */
uint16_t threads_per_core;/* threads per core required by job */
uint16_t ntasks_per_node;/* number of tasks to invoke on each node */
uint16_t ntasks_per_socket;/* number of tasks to invoke on
* each socket */
uint16_t ntasks_per_core;/* number of tasks to invoke on each core */
uint16_t ntasks_per_board;/* number of tasks to invoke on each board */
uint16_t ntasks_per_tres;/* number of tasks that can access each gpu */
uint16_t pn_min_cpus; /* minimum # CPUs per node, default=0 */
uint64_t pn_min_memory; /* minimum real memory per node OR
* real memory per CPU | MEM_PER_CPU,
* default=0 (no limit) */
uint32_t pn_min_tmp_disk;/* minimum tmp disk per node,
* default=0 */
char *req_context; /* requested selinux context */
uint32_t req_switch; /* Maximum number of switches */
uint16_t segment_size; /* segment_size */
char *selinux_context; /* used internally in the slurmctld,
DON'T PACK */
char *std_err; /* pathname of stderr */
char *std_in; /* pathname of stdin */
char *std_out; /* pathname of stdout */
uint64_t *tres_req_cnt; /* used internally in the slurmctld,
DON'T PACK */
uint32_t wait4switch; /* Maximum time to wait for minimum switches */
char *wckey; /* wckey for job */
uint16_t x11; /* --x11 flags */
char *x11_magic_cookie; /* automatically stolen from submit node */
char *x11_target; /* target hostname, or unix socket if port == 0 */
uint16_t x11_target_port; /* target tcp port, 6000 + the display number */
} job_desc_msg_t;
typedef struct job_info {
char *account; /* charge to specified account */
time_t accrue_time; /* time job is eligible for running */
char *admin_comment; /* administrator's arbitrary comment */
char *alloc_node; /* local node making resource alloc */
uint32_t alloc_sid; /* local sid making resource alloc */
bitstr_t *array_bitmap; /* NOTE: set on unpack */
uint32_t array_job_id; /* job_id of a job array or 0 if N/A */
uint32_t array_task_id; /* task_id of a job array */
uint32_t array_max_tasks; /* Maximum number of running tasks */
char *array_task_str; /* string expression of task IDs in this record */
uint32_t assoc_id; /* association id for job */
char *batch_features; /* features required for batch script's node */
uint16_t batch_flag; /* 1 if batch: queued job with script */
char *batch_host; /* name of host running batch script */
uint64_t bitflags; /* Various job flags */
uint16_t boards_per_node; /* boards per node required by job */
char *burst_buffer; /* burst buffer specifications */
char *burst_buffer_state; /* burst buffer state info */
char *cluster; /* name of cluster that the job is on */
char *cluster_features; /* comma separated list of required cluster
* features */
char *command; /* command to be executed, built from submitted
* job's argv */
char *comment; /* arbitrary comment */
char *container; /* OCI Container bundle path */
char *container_id; /* OCI Container ID */
uint16_t contiguous; /* 1 if job requires contiguous nodes */
uint16_t core_spec; /* specialized core count */
uint16_t cores_per_socket; /* cores per socket required by job */
double billable_tres; /* billable TRES cache. updated upon resize */
uint16_t cpus_per_task; /* number of processors required for
* each task */
uint32_t cpu_freq_min; /* Minimum cpu frequency */
uint32_t cpu_freq_max; /* Maximum cpu frequency */
uint32_t cpu_freq_gov; /* cpu frequency governor */
char *cpus_per_tres; /* semicolon delimited list of TRES=# values */
char *cronspec; /* cron time specification (scrontab jobs) */
time_t deadline; /* deadline */
uint32_t delay_boot; /* delay boot for desired node state */
char *dependency; /* synchronize job execution with other jobs */
uint32_t derived_ec; /* highest exit code of all job steps */
time_t eligible_time; /* time job is eligible for running */
time_t end_time; /* time of termination, actual or expected */
char *exc_nodes; /* comma separated list of excluded nodes */
int32_t *exc_node_inx; /* excluded list index pairs into node_table:
* start_range_1, end_range_1,
* start_range_2, .., -1 */
uint32_t exit_code; /* exit code for job (status from wait call) */
char *extra; /* Arbitrary string */
char *failed_node; /* if set, node that caused job to fail */
char *features; /* comma separated list of required features */
char *fed_origin_str; /* Origin cluster's name */
uint64_t fed_siblings_active; /* bitmap of active fed sibling ids */
char *fed_siblings_active_str; /* string of active sibling names */
uint64_t fed_siblings_viable; /* bitmap of viable fed sibling ids */
char *fed_siblings_viable_str; /* string of viable sibling names */
uint32_t gres_detail_cnt; /* Count of gres_detail_str records,
* one per allocated node */
char **gres_detail_str; /* Details of GRES count/index alloc per node */
char *gres_total; /* Total count of gres used with names */
uint32_t group_id; /* group job submitted as */
uint32_t het_job_id; /* job ID of hetjob leader */
char *het_job_id_set; /* job IDs for all components */
uint32_t het_job_offset; /* HetJob component offset from leader */
uint32_t job_id; /* job ID */
job_resources_t *job_resrcs; /* opaque data type, job resources */
char *job_size_str;
uint32_t job_state; /* state of the job, see enum job_states */
time_t last_sched_eval; /* last time job was evaluated for scheduling */
char *licenses; /* licenses required by the job */
uint16_t mail_type; /* see MAIL_JOB_ definitions above */
char *mail_user; /* user to receive notification */
uint32_t max_cpus; /* maximum number of cpus usable by job */
uint32_t max_nodes; /* maximum number of nodes usable by job */
char *mcs_label; /* mcs_label if mcs plugin in use */
char *mem_per_tres; /* semicolon delimited list of TRES=# values */
char *name; /* name of the job */
char *network; /* network specification */
char *nodes; /* list of nodes allocated to job */
uint32_t nice; /* requested priority change */
int32_t *node_inx; /* list index pairs into node_table for *nodes:
* start_range_1, end_range_1,
* start_range_2, .., -1 */
uint16_t ntasks_per_core;/* number of tasks to invoke on each core */
uint16_t ntasks_per_tres;/* number of tasks that can access each gpu */
uint16_t ntasks_per_node;/* number of tasks to invoke on each node */
uint16_t ntasks_per_socket;/* number of tasks to invoke on each socket*/
uint16_t ntasks_per_board; /* number of tasks to invoke on each board */
uint32_t num_cpus; /* minimum number of cpus required by job */
uint32_t num_nodes; /* minimum number of nodes required by job */
uint32_t num_tasks; /* requested task count */
uint16_t oom_kill_step; /* kill whole step in case of task oom */
char *partition; /* name of assigned partition */
char *prefer; /* comma separated list of soft features */
uint64_t pn_min_memory; /* minimum real memory per node, default=0 */
uint16_t pn_min_cpus; /* minimum # CPUs per node, default=0 */
uint32_t pn_min_tmp_disk; /* minimum tmp disk per node, default=0 */
time_t preempt_time; /* preemption signal time */
time_t preemptable_time;/* job becomes preemptable from
* PreemptExemptTime */
time_t pre_sus_time; /* time job ran prior to last suspend */
uint32_t priority; /* relative priority of the job,
* 0=held, 1=required nodes DOWN/DRAINED */
uint32_t *priority_array; /* partition based priority */
char *priority_array_names; /* partition order of priority_array */
uint32_t profile; /* Level of acct_gather_profile {all | none} */
char *qos; /* Quality of Service */
uint8_t reboot; /* node reboot requested before start */
char *req_nodes; /* comma separated list of required nodes */
int32_t *req_node_inx; /* required list index pairs into node_table:
* start_range_1, end_range_1,
* start_range_2, .., -1 */
uint32_t req_switch; /* Maximum number of switches */
uint16_t requeue; /* enable or disable job requeue option */
time_t resize_time; /* time of latest size change */
uint16_t restart_cnt; /* count of job restarts */
char *resv_name; /* reservation name */
char *resv_ports; /* Reserved MPI ports (as hostlist str) */
char *sched_nodes; /* list of nodes scheduled to be used for job */
char *selinux_context;
uint16_t shared; /* 1 if job can share nodes with other jobs */
uint32_t site_factor; /* factor to consider in priority */
uint16_t sockets_per_board;/* sockets per board required by job */
uint16_t sockets_per_node; /* sockets per node required by job */
time_t start_time; /* time execution begins, actual or expected */
uint16_t start_protocol_ver; /* Slurm version step was started with
* either srun or the lowest slurmd version
* it is talking to */
char *state_desc; /* optional details for state_reason */
uint32_t state_reason; /* reason job still pending or failed, see
* slurm.h:enum job_state_reason */
char *std_err; /* pathname of job's stderr file */
char *std_in; /* pathname of job's stdin file */
char *std_out; /* pathname of job's stdout file */
time_t submit_time; /* time of job submission */
time_t suspend_time; /* time job last suspended or resumed */
char *system_comment; /* slurmctld's arbitrary comment */
uint32_t time_limit; /* maximum run time in minutes or INFINITE */
uint32_t time_min; /* minimum run time in minutes or INFINITE */
uint16_t threads_per_core; /* threads per core required by job */
char *tres_bind; /* Task to TRES binding directives */
char *tres_freq; /* TRES frequency directives */
char *tres_per_job; /* semicolon delimited list of TRES=# values */
char *tres_per_node; /* semicolon delimited list of TRES=# values */
char *tres_per_socket; /* semicolon delimited list of TRES=# values */
char *tres_per_task; /* semicolon delimited list of TRES=# values */
char *tres_req_str; /* tres requested in the job */
char *tres_alloc_str; /* tres used in the job */
uint32_t user_id; /* user the job runs as */
char *user_name; /* user_name or null. not always set, but
* accurate if set (and can avoid a local
* lookup call) */
uint32_t wait4switch; /* Maximum time to wait for minimum switches */
char *wckey; /* wckey for job */
char *work_dir; /* pathname of working directory */
} slurm_job_info_t;
typedef slurm_job_info_t job_info_t;
typedef struct {
uint32_t nice;
double priority_age;
double priority_assoc;
double priority_fs;
double priority_js;
double priority_part;
double priority_qos;
uint32_t priority_site;
double *priority_tres;/* tres priorities with weights applied. */
uint32_t tres_cnt; /* number of configured tres' on system. */
char **tres_names; /* packed as assoc_mgr_tres_names[] */
double *tres_weights; /* PriorityWeightTRES weights as an array */
} priority_factors_t;
typedef struct priority_factors_object {
char *account;
char *cluster_name; /* Cluster name ONLY set in federation */
double direct_prio; /* Manually set priority. If it is set prio_factors
* will be NULL */
uint32_t job_id;
char *partition;
priority_factors_t *prio_factors;
char *qos;
uint32_t user_id;
} priority_factors_object_t;
typedef struct priority_factors_response_msg {
list_t *priority_factors_list; /* priority_factors_object_t list */
} priority_factors_response_msg_t;
typedef struct job_info_msg {
time_t last_backfill; /* time of late backfill run */
time_t last_update; /* time of latest info */
uint32_t record_count; /* number of records */
slurm_job_info_t *job_array; /* the job records */
} job_info_msg_t;
typedef struct listjobs_info {
uint32_t job_id;
} listjobs_info_t;
typedef struct listpids_info {
uint32_t global_task_id;
uint32_t job_id;
uint32_t local_task_id;
pid_t pid;
char *step_id;
} listpids_info_t;
typedef struct liststeps_info {
uint32_t job_id;
char *step_id;
} liststeps_info_t;
typedef struct {
uint32_t job_id;
uint32_t array_job_id;
uint32_t array_task_id;
bitstr_t *array_task_id_bitmap;
/* state of the job; see enum job_states and JOB_* macros */
uint32_t het_job_id; /* JobId of hetjob leader */
uint32_t state;
} job_state_response_job_t;
typedef struct {
uint32_t jobs_count;
job_state_response_job_t *jobs;
} job_state_response_msg_t;
typedef struct step_update_request_msg {
uint32_t job_id;
uint32_t step_id;
uint32_t time_limit; /* In minutes */
} step_update_request_msg_t;
typedef struct suspend_exc_update_msg {
char *update_str;
update_mode_t mode;
} suspend_exc_update_msg_t;
typedef struct {
char *node_list; /* nodelist corresponding to task layout */
uint16_t *cpus_per_node; /* cpus per node */
uint32_t *cpu_count_reps; /* how many nodes have same cpu count */
uint32_t num_hosts; /* number of hosts we have */
uint32_t num_tasks; /* number of tasks to distribute across these cpus*/
uint16_t *cpus_per_task; /* number of cpus per task */
uint32_t *cpus_task_reps; /* how many nodes have same per task count */
uint32_t task_dist; /* type of distribution we are using */
uint16_t plane_size; /* plane size (only needed for plane distribution*/
} slurm_step_layout_req_t;
typedef struct slurm_step_layout {
uint16_t *cpt_compact_array; /* Compressed per-node cpus_per_task.
* Index with slurm_get_rep_count_inx() */
uint32_t cpt_compact_cnt; /* number of elements in cpt_compact arrays */
uint32_t *cpt_compact_reps; /* number of consecutive nodes on which a
* value in cpt_compact_array is
* duplicated */
char *front_end; /* If a front-end architecture, the name of
* of the node running all tasks,
* NULL otherwise */
slurm_node_alias_addrs_t *alias_addrs;
uint32_t node_cnt; /* node count */
char *node_list; /* list of nodes in step */
uint16_t plane_size; /* plane size when task_dist =
* SLURM_DIST_PLANE */
uint16_t start_protocol_ver; /* Slurm version step was started with
* either srun or the lowest slurmd version
* it is talking to */
/* Array of length "node_cnt". Each element of the array
* is the number of tasks assigned to the corresponding node */
uint16_t *tasks;
uint32_t task_cnt; /* total number of tasks in the step */
uint32_t task_dist; /* see enum task_dist_state */
/* Array (of length "node_cnt") of task ID arrays. The length
* of each subarray is designated by the corresponding value in
* the tasks array. */
uint32_t **tids; /* host id => task id mapping */
} slurm_step_layout_t;
typedef struct slurm_step_id_msg {
sluid_t sluid;
uint32_t job_id;
uint32_t step_het_comp;
uint32_t step_id;
} slurm_step_id_t;
typedef struct slurm_step_io_fds {
struct {
int fd;
uint32_t taskid;
uint32_t nodeid;
} input, out, err;
} slurm_step_io_fds_t;
#define SLURM_STEP_IO_FDS_INITIALIZER {{0, (uint32_t)-1, (uint32_t)-1}, \
{1, (uint32_t)-1, (uint32_t)-1}, \
{2, (uint32_t)-1, (uint32_t)-1}}
typedef struct launch_tasks_response_msg {
uint32_t return_code;
char *node_name;
uint32_t srun_node_id;
uint32_t count_of_pids;
uint32_t *local_pids;
slurm_step_id_t step_id;
uint32_t *task_ids; /* array of length count_of_pids */
} launch_tasks_response_msg_t;
typedef struct task_ext_msg {
uint32_t num_tasks;
uint32_t *task_id_list;
uint32_t return_code;
slurm_step_id_t step_id;
} task_exit_msg_t;
typedef struct {
uint32_t job_id; /* slurm job_id */
uint32_t flags; /* flags */
uint16_t port; /* target TCP port */
char *target; /* target host or UNIX socket */
} net_forward_msg_t;
typedef struct srun_ping_msg {
uint32_t job_id; /* slurm job_id */
} srun_ping_msg_t;
typedef slurm_step_id_t srun_job_complete_msg_t;
typedef struct srun_timeout_msg {
slurm_step_id_t step_id;
time_t timeout; /* when job scheduled to be killed */
} srun_timeout_msg_t;
typedef struct srun_user_msg {
uint32_t job_id; /* slurm job_id */
char *msg; /* message to user's srun */
} srun_user_msg_t;
typedef struct srun_node_fail_msg {
char *nodelist; /* name of failed node(s) */
slurm_step_id_t step_id;
} srun_node_fail_msg_t;
typedef struct srun_step_missing_msg {
char *nodelist; /* name of node(s) lacking this step */
slurm_step_id_t step_id;
} srun_step_missing_msg_t;
enum suspend_opts {
SUSPEND_JOB, /* Suspend a job now */
RESUME_JOB /* Resume a job now */
};
/* NOTE: Set either job_id_str (NULL by default) or job_id */
typedef struct suspend_msg {
uint16_t op; /* suspend operation, see enum suspend_opts */
uint32_t job_id; /* slurm job ID (number) */
char * job_id_str; /* slurm job ID (string) */
} suspend_msg_t;
/* NOTE: Set either job_id_str (NULL by default) or job_id */
typedef struct top_job_msg {
uint16_t op; /* suspend operation, see enum suspend_opts */
uint32_t job_id; /* slurm job ID (number) */
char * job_id_str; /* slurm job ID (string) */
} top_job_msg_t;
typedef struct {
char *alias_list; /* node name/address/hostname aliases */
uint32_t argc;
char **argv;
uint32_t envc;
char **env;
char *container; /* OCI Container bundle path */
char *cwd;
uint32_t msg_timeout; /* timeout set for sending message */
uint16_t ntasks_per_board;/* number of tasks to invoke on each board */
uint16_t ntasks_per_core; /* number of tasks to invoke on each core */
uint16_t ntasks_per_tres;/* number of tasks that can access each gpu */
uint16_t ntasks_per_socket;/* number of tasks to invoke on
* each socket */
/* I/O handling */
bool buffered_stdio;
bool labelio;
char *remote_output_filename;
char *remote_error_filename;
char *remote_input_filename;
slurm_step_io_fds_t local_fds;
bool multi_prog;
bool no_alloc;
uint32_t slurmd_debug; /* remote slurmd debug level */
uint32_t het_job_node_offset; /* Hetjob node offset or NO_VAL */
uint32_t het_job_id; /* Hetjob ID or NO_VAL */
uint32_t het_job_nnodes;/* total task count for entire hetjob */
uint32_t het_job_ntasks;/* total task count for entire hetjob */
uint32_t het_job_step_cnt;/* total step count for entire hetjob */
uint16_t *het_job_task_cnts; /* Number of tasks on each node in hetjob */
uint32_t **het_job_tids; /* Task IDs on each node in hetjob */
uint32_t *het_job_tid_offsets;/* map of tasks (by id) to originating
* hetjob */
uint32_t het_job_offset;/* Hetjob offset or NO_VAL */
uint32_t het_job_task_offset; /* Hetjob task offset or NO_VAL */
char *het_job_node_list; /* Hetjob step node list */
bool parallel_debug;
uint32_t profile; /* Level of acct_gather_profile {all | none} */
char *task_prolog;
char *task_epilog;
uint16_t cpu_bind_type; /* use cpu_bind_type_t */
char *cpu_bind;
uint32_t cpu_freq_min; /* Minimum cpu frequency */
uint32_t cpu_freq_max; /* Maximum cpu frequency */
uint32_t cpu_freq_gov; /* cpu frequency governor */
uint16_t mem_bind_type; /* use mem_bind_type_t */
char *mem_bind;
uint16_t accel_bind_type; /* --accel-bind= */
uint16_t max_sockets;
uint16_t max_cores;
uint16_t max_threads;
uint16_t cpus_per_task;
uint16_t *cpt_compact_array; /* Compressed per-node cpus_per_task.
* Index with slurm_get_rep_count_inx() */
uint32_t cpt_compact_cnt; /* number of elements in cpt_compact arrays */
uint32_t *cpt_compact_reps; /* number of consecutive nodes on which a
* value in cpt_compact_array is
* duplicated */
uint16_t threads_per_core;
uint32_t task_dist;
uint16_t tree_width;
bool preserve_env;
char *mpi_plugin_name;
uint8_t open_mode;
char *acctg_freq;
bool pty;
char **spank_job_env; /* environment variables for job prolog/epilog
* scripts as set by SPANK plugins */
uint32_t spank_job_env_size; /* element count in spank_env */
char *tres_bind;
char *tres_freq;
uint16_t oom_kill_step;
} slurm_step_launch_params_t;
typedef struct {
void (*step_complete)(srun_job_complete_msg_t *);
void (*step_signal)(int);
void (*step_timeout)(srun_timeout_msg_t *);
void (*task_start)(launch_tasks_response_msg_t *);
void (*task_finish)(task_exit_msg_t *);
} slurm_step_launch_callbacks_t;
typedef struct {
void (*job_complete)(srun_job_complete_msg_t *);
void (*timeout)(srun_timeout_msg_t *);
void (*user_msg)(srun_user_msg_t *);
void (*node_fail)(srun_node_fail_msg_t *);
void (*job_suspend)(suspend_msg_t *);
} slurm_allocation_callbacks_t;
typedef struct {
void (*acct_full)();
void (*dbd_fail)();
void (*dbd_resumed)();
void (*db_fail)();
void (*db_resumed)();
} slurm_trigger_callbacks_t;
typedef struct {
uint32_t array_job_id; /* job_id of a job array or 0 if N/A */
uint32_t array_task_id; /* task_id of a job array */
char *cluster; /* cluster that the step is running on. */
char *container; /* OCI container bundle path */
char *container_id; /* OCI container ID */
uint32_t cpu_freq_min; /* Minimum cpu frequency */
uint32_t cpu_freq_max; /* Maximum cpu frequency */
uint32_t cpu_freq_gov; /* cpu frequency governor */
char *cpus_per_tres; /* comma delimited list of TRES=# values */
char *mem_per_tres; /* comma delimited list of TRES=# values */
char *name; /* name of job step */
char *network; /* network specs for job step */
char *nodes; /* list of nodes allocated to job_step */
int32_t *node_inx; /* list index pairs into node_table for *nodes:
* start_range_1, end_range_1,
* start_range_2, .., -1 */
uint32_t num_cpus; /* how many cpus are being used by step */
uint32_t num_tasks; /* number of tasks */
char *partition; /* name of assigned partition */
char *resv_ports; /* ports allocated for MPI */
time_t run_time; /* net run time (factor out time suspended) */
char *srun_host; /* host of srun command */
uint32_t srun_pid; /* PID of srun command */
time_t start_time; /* step start time */
uint16_t start_protocol_ver; /* Slurm version step was started with
* either srun or the lowest slurmd version
* it is talking to */
uint32_t state; /* state of the step, see enum job_states */
slurm_step_id_t step_id;
char *submit_line; /* The command issued with all it's options in a
* string */
uint32_t task_dist; /* see enum task_dist_state */
uint32_t time_limit; /* step time limit */
char *tres_alloc_str; /* tres used in the job */
char *tres_bind; /* Task to TRES binding directives */
char *tres_freq; /* TRES frequency directives */
char *tres_per_step; /* comma delimited list of TRES=# values */
char *tres_per_node; /* comma delimited list of TRES=# values */
char *tres_per_socket; /* comma delimited list of TRES=# values */
char *tres_per_task; /* comma delimited list of TRES=# values */
uint32_t user_id; /* user the job runs as */
} job_step_info_t;
typedef struct {
uint32_t job_id;
char *stepmgr;
} stepmgr_job_info_t;
typedef struct job_step_info_response_msg {
time_t last_update; /* time of latest info */
uint32_t job_step_count; /* number of records */
job_step_info_t *job_steps; /* the job step records */
list_t *stepmgr_jobs;
} job_step_info_response_msg_t;
typedef struct {
char *node_name;
uint32_t *pid;
uint32_t pid_cnt;
} job_step_pids_t;
typedef struct {
list_t *pid_list; /* list of job_step_pids_t *'s */
slurm_step_id_t step_id;
} job_step_pids_response_msg_t;
typedef struct {
jobacctinfo_t *jobacct;
uint32_t num_tasks;
uint32_t return_code;
job_step_pids_t *step_pids;
} job_step_stat_t;
typedef struct {
list_t *stats_list; /* list of job_step_stat_t *'s */
slurm_step_id_t step_id;
} job_step_stat_response_msg_t;
typedef struct node_info {
char *arch; /* computer architecture */
char *bcast_address; /* BcastAddr (optional) */
uint16_t boards; /* total number of boards per node */
time_t boot_time; /* time of node boot */
char *cluster_name; /* Cluster name ONLY set in federation */
uint16_t cores; /* number of cores per socket */
uint16_t core_spec_cnt; /* number of specialized cores on node */
uint32_t cpu_bind; /* Default task binding */
uint32_t cpu_load; /* CPU load * 100 */
uint64_t free_mem; /* free memory in MiB */
uint16_t cpus; /* configured count of cpus running on
* the node */
uint16_t cpus_efctv; /* count of effective cpus on the node.
i.e cpus minus specialized cpus*/
char *cpu_spec_list; /* node's specialized cpus */
acct_gather_energy_t *energy; /* energy data */
char *extra; /* arbitrary sting */
char *features; /* list of a node's available features */
char *features_act; /* list of a node's current active features,
* Same as "features" if NULL */
char *gres; /* list of a node's generic resources */
char *gres_drain; /* list of drained GRES */
char *gres_used; /* list of GRES in current use */
char *instance_id; /* cloud instance id */
char *instance_type; /* cloud instance type */
time_t last_busy; /* time node was last busy (i.e. no jobs) */
char *mcs_label; /* mcs label if mcs plugin in use */
uint64_t mem_spec_limit; /* MB memory limit for specialization */
char *name; /* node name to slurm */
uint32_t next_state; /* state after reboot (enum node_states) */
char *node_addr; /* communication name (optional) */
char *node_hostname; /* node's hostname (optional) */
uint32_t node_state; /* see enum node_states */
char *os; /* operating system currently running */
uint32_t owner; /* User allowed to use this node or NO_VAL */
char *partitions; /* Comma separated list of partitions containing
* this node, NOT supplied by slurmctld, but
* populated by scontrol */
uint16_t port; /* TCP port number of the slurmd */
uint64_t real_memory; /* configured MB of real memory on the node */
uint16_t res_cores_per_gpu; /* number of cores per GPU to allow
* to only GPU jobs */
char *gpu_spec; /* node's cores reserved for GPU jobs */
char *comment; /* arbitrary comment */
char *reason; /* reason for node being DOWN or DRAINING */
time_t reason_time; /* Time stamp when reason was set, ignore if
* no reason is set. */
uint32_t reason_uid; /* User that set the reason, ignore if
* no reason is set. */
time_t resume_after; /* automatically resume DOWN or DRAINED node at
* this point in time */
char *resv_name; /* If node is in a reservation this is
* the name of the reservation */
dynamic_plugin_data_t *select_nodeinfo; /* opaque data structure,
* use
* slurm_get_select_nodeinfo()
* to access contents */
time_t slurmd_start_time;/* time of slurmd startup */
uint16_t sockets; /* total number of sockets per node */
uint16_t threads; /* number of threads per core */
uint32_t tmp_disk; /* configured MB of total disk in TMP_FS */
uint32_t weight; /* arbitrary priority of node for scheduling */
char *tres_fmt_str; /* str representing configured TRES on node */
char *version; /* Slurm version number */
} node_info_t;
typedef struct node_info_msg {
time_t last_update; /* time of latest info */
uint32_t record_count; /* number of records */
node_info_t *node_array; /* the node records */
} node_info_msg_t;
typedef struct front_end_info {
char *allow_groups; /* allowed group string */
char *allow_users; /* allowed user string */
time_t boot_time; /* Time of node boot,
* computed from up_time */
char *deny_groups; /* denied group string */
char *deny_users; /* denied user string */
char *name; /* node name */
uint32_t node_state; /* see enum node_states */
char *reason; /* reason for node being DOWN or
* DRAINING */
time_t reason_time; /* Time stamp when reason was set,
* ignore if no reason is set. */
uint32_t reason_uid; /* User that set the reason,
* ignore if no reason is set. */
time_t slurmd_start_time; /* Time of slurmd startup */
char *version; /* Slurm version number */
} front_end_info_t;
typedef struct front_end_info_msg {
time_t last_update; /* time of latest info */
uint32_t record_count; /* number of records */
front_end_info_t *front_end_array; /* the front_end records */
} front_end_info_msg_t;
typedef struct topo_info {
uint16_t level; /* level in hierarchy, leaf=0 */
uint32_t link_speed; /* link speed, arbitrary units */
char *name; /* switch name */
char *nodes; /* name if direct descendent nodes */
char *switches; /* name if direct descendent switches */
} topo_info_t;
typedef struct topo_info_response_msg {
uint32_t record_count; /* number of records */
topo_info_t *topo_array; /* the switch topology records */
dynamic_plugin_data_t *topo_info;
} topo_info_response_msg_t;
typedef struct job_alloc_info_msg {
uint32_t job_id; /* job ID */
char *req_cluster; /* requesting cluster */
} job_alloc_info_msg_t;
#define SLURM_SELECTED_STEP_INITIALIZER \
{ NULL, NO_VAL, NO_VAL, { NO_VAL, NO_VAL, NO_VAL } }
typedef struct {
bitstr_t *array_bitmap; /* Set with slurm_array_str2bitmap(). */
uint32_t array_task_id; /* task_id of a job array or NO_VAL */
uint32_t het_job_offset; /* het_job_offset or NO_VAL */
slurm_step_id_t step_id;
} slurm_selected_step_t;
typedef slurm_selected_step_t step_alloc_info_msg_t;
typedef struct acct_gather_node_resp_msg {
acct_gather_energy_t *energy;
char *node_name; /* node name */
uint16_t sensor_cnt;
} acct_gather_node_resp_msg_t;
typedef struct acct_gather_energy_req_msg {
uint16_t context_id;
uint16_t delta;
} acct_gather_energy_req_msg_t;
#define JOB_DEF_CPU_PER_GPU 0x0001
#define JOB_DEF_MEM_PER_GPU 0x0002
typedef struct job_defaults {
uint16_t type; /* See JOB_DEF_* above */
uint64_t value; /* Value */
} job_defaults_t;
/* Current partition state information and used to set partition options
* using slurm_update_partition(). */
#define PART_FLAG_DEFAULT SLURM_BIT(0) /* Set if default partition */
#define PART_FLAG_HIDDEN SLURM_BIT(1) /* Set if partition is hidden */
#define PART_FLAG_NO_ROOT SLURM_BIT(2) /* Set if user root jobs disabled */
#define PART_FLAG_ROOT_ONLY SLURM_BIT(3) /* Set if only root can submit jobs */
#define PART_FLAG_REQ_RESV SLURM_BIT(4) /* Set if reservation is required */
#define PART_FLAG_LLN SLURM_BIT(5) /* Set if least loaded node selection
* is desired */
#define PART_FLAG_EXCLUSIVE_USER SLURM_BIT(6)/* Set if nodes allocated exclusively
* by user */
#define PART_FLAG_PDOI SLURM_BIT(7) /* Set if nodes POWER_DOWN on IDLE,
* after running jobs */
/* Used with slurm_update_partition() to clear flags associated with existing
* partitions. For example, if a partition is currently hidden and you want
* to make it visible then set flags to PART_FLAG_HIDDEN_CLR and call
* slurm_update_partition(). */
#define PART_FLAG_DEFAULT_CLR SLURM_BIT(8) /* Clear DEFAULT partition flag */
#define PART_FLAG_HIDDEN_CLR SLURM_BIT(9) /* Clear HIDDEN partition flag */
#define PART_FLAG_NO_ROOT_CLR SLURM_BIT(10) /* Clear NO_ROOT partition flag */
#define PART_FLAG_ROOT_ONLY_CLR SLURM_BIT(11) /* Clear ROOT_ONLY partition flag */
#define PART_FLAG_REQ_RESV_CLR SLURM_BIT(12) /* Clear RES_REQ partition flag */
#define PART_FLAG_LLN_CLR SLURM_BIT(13) /* Clear LLN partition flag */
#define PART_FLAG_EXC_USER_CLR SLURM_BIT(14) /* Clear EXCLUSIVE_USER flag */
#define PART_FLAG_PDOI_CLR SLURM_BIT(15) /* Clear PDOI partition flag */
#define PART_FLAG_EXCLUSIVE_TOPO SLURM_BIT(16)/* Set if Topo allocated exclusively */
#define PART_FLAG_EXC_TOPO_CLR SLURM_BIT(17) /* Clear EXCLUSIVE_TOPO flag */
#define PART_FLAG_SCHED_FAILED SLURM_BIT(18) /* Partition failed to schedule
* job */
#define PART_FLAG_SCHED_CLEARED SLURM_BIT(19) /* Partition's nodes have been
* cleared from available after
* failed to schedule job */
typedef struct partition_info {
char *allow_alloc_nodes;/* list names of allowed allocating
* nodes */
char *allow_accounts; /* comma delimited list of accounts,
* null indicates all */
char *allow_groups; /* comma delimited list of groups,
* null indicates all */
char *allow_qos; /* comma delimited list of qos,
* null indicates all */
char *alternate; /* name of alternate partition */
char *billing_weights_str;/* per TRES billing weights string */
char *cluster_name; /* Cluster name ONLY set in federation */
uint16_t cr_type; /* see CR_* values */
uint32_t cpu_bind; /* Default task binding */
uint64_t def_mem_per_cpu; /* default MB memory per allocated CPU */
uint32_t default_time; /* minutes, NO_VAL or INFINITE */
char *deny_accounts; /* comma delimited list of denied accounts */
char *deny_qos; /* comma delimited list of denied qos */
uint32_t flags; /* see PART_FLAG_* above */
uint32_t grace_time; /* preemption grace time in seconds */
list_t *job_defaults_list; /* List of job_defaults_t elements */
char *job_defaults_str; /* String of job defaults,
* used only for partition update RPC */
uint32_t max_cpus_per_node; /* maximum allocated CPUs per node */
uint32_t max_cpus_per_socket; /* maximum allocated CPUs per socket */
uint64_t max_mem_per_cpu; /* maximum MB memory per allocated CPU */
uint32_t max_nodes; /* per job or INFINITE */
uint16_t max_share; /* number of jobs to gang schedule */
uint32_t max_time; /* minutes or INFINITE */
uint32_t min_nodes; /* per job */
char *name; /* name of the partition */
int32_t *node_inx; /* list index pairs into node_table:
* start_range_1, end_range_1,
* start_range_2, .., -1 */
char *nodes; /* list names of nodes in partition */
char *nodesets; /* list of nodesets used by partition */
uint16_t over_time_limit; /* job's time limit can be exceeded by this
* number of minutes before cancellation */
uint16_t preempt_mode; /* See PREEMPT_MODE_* in slurm/slurm.h */
uint16_t priority_job_factor; /* job priority weight factor */
uint16_t priority_tier; /* tier for scheduling and preemption */
char *qos_char; /* The partition QOS name */
uint16_t resume_timeout; /* time required in order to perform a node
* resume operation */
uint16_t state_up; /* see PARTITION_ states above */
uint32_t suspend_time; /* node idle for this long before power save
* mode */
uint16_t suspend_timeout; /* time required in order to perform a node
* suspend operation */
uint32_t total_cpus; /* total number of cpus in the partition */
uint32_t total_nodes; /* total number of nodes in the partition */
char *tres_fmt_str; /* str of configured TRES in partition */
} partition_info_t;
typedef struct delete_partition_msg {
char *name; /* name of partition to be delete */
} delete_part_msg_t;
typedef struct resource_allocation_response_msg {
char *account; /* allocation account */
uint32_t job_id; /* assigned job id */
char *alias_list; /* node name/address/hostname aliases */
char *batch_host; /* host executing batch script */
uint32_t cpu_freq_min; /* Minimum cpu frequency */
uint32_t cpu_freq_max; /* Maximum cpu frequency */
uint32_t cpu_freq_gov; /* cpu frequency governor */
uint16_t *cpus_per_node;/* cpus per node */
uint32_t *cpu_count_reps;/* how many nodes have same cpu count */
uint32_t env_size; /* element count in environment */
char **environment; /* environment variables to set for job,
* name=value pairs, one per line */
uint32_t error_code; /* error code for warning message */
gid_t gid; /* resolved group id of job */
char *group_name; /* resolved group name of job */
char *job_submit_user_msg;/* job_submit plugin user_msg */
slurm_addr_t *node_addr; /* network addresses */
uint32_t node_cnt; /* count of nodes */
char *node_list; /* assigned list of nodes */
uint16_t ntasks_per_board;/* number of tasks to invoke on each board */
uint16_t ntasks_per_core; /* number of tasks to invoke on each core */
uint16_t ntasks_per_tres;/* number of tasks that can access each gpu */
uint16_t ntasks_per_socket;/* number of tasks to invoke on
* each socket */
uint32_t num_cpu_groups;/* size of cpus_per_node and cpu_count_reps */
char *partition; /* name of partition used to run job */
uint64_t pn_min_memory; /* minimum real memory per node OR
* real memory per CPU | MEM_PER_CPU,
* default=0 (no limit) */
char *qos; /* allocation qos */
char *resv_name; /* allocation reservation */
char *tres_per_node; /* comma delimited list of TRES=# values */
uid_t uid; /* resolved user id of job */
char *user_name; /* resolved user name of job */
void *working_cluster_rec; /* Cluster to direct remaining messages to.
* slurmdb_cluster_rec_t* because slurm.h
* doesn't know about slurmdb.h. */
} resource_allocation_response_msg_t;
typedef struct partition_info_msg {
time_t last_update; /* time of latest info */
uint32_t record_count; /* number of records */
partition_info_t *partition_array; /* the partition records */
} partition_info_msg_t;
typedef struct will_run_response_msg {
uint32_t job_id; /* ID of job to start */
char *job_submit_user_msg; /* job submit plugin user_msg */
char *node_list; /* nodes where job will start */
char *part_name; /* partition where job will start */
list_t *preemptee_job_id; /* jobs preempted to start this job */
uint32_t proc_cnt; /* CPUs allocated to job at start */
time_t start_time; /* time when job will start */
double sys_usage_per; /* System usage percentage */
} will_run_response_msg_t;
/*********************************/
/*
* Resource reservation data structures.
* Create, show, modify and delete functions are required
*/
#define RESERVE_FLAG_MAINT SLURM_BIT(0) /* Set MAINT flag */
#define RESERVE_FLAG_NO_MAINT SLURM_BIT(1) /* Clear MAINT flag */
#define RESERVE_FLAG_DAILY SLURM_BIT(2) /* Set DAILY flag */
#define RESERVE_FLAG_NO_DAILY SLURM_BIT(3) /* Clear DAILY flag */
#define RESERVE_FLAG_WEEKLY SLURM_BIT(4) /* Set WEEKLY flag */
#define RESERVE_FLAG_NO_WEEKLY SLURM_BIT(5) /* Clear WEEKLY flag */
#define RESERVE_FLAG_IGN_JOBS SLURM_BIT(6) /* Ignore running jobs */
#define RESERVE_FLAG_NO_IGN_JOB SLURM_BIT(7) /* Clear ignore running
* jobs flag */
#define RESERVE_FLAG_ANY_NODES SLURM_BIT(8) /* Use any compute nodes */
#define RESERVE_FLAG_NO_ANY_NODES SLURM_BIT(9) /* Clear any compute
* node flag */
#define RESERVE_FLAG_STATIC SLURM_BIT(10) /* Static node allocation */
#define RESERVE_FLAG_NO_STATIC SLURM_BIT(11) /* Clear static node
* allocation */
#define RESERVE_FLAG_PART_NODES SLURM_BIT(12) /* Use partition nodes only */
#define RESERVE_FLAG_NO_PART_NODES SLURM_BIT(13) /* Clear partition
* nodes only flag */
#define RESERVE_FLAG_OVERLAP SLURM_BIT(14) /* Permit to overlap others */
#define RESERVE_FLAG_SPEC_NODES SLURM_BIT(15) /* Contains specific nodes */
/* SLURM_BIT(16) Available 2 versions after 23.11 */
#define RESERVE_FLAG_TIME_FLOAT SLURM_BIT(17) /* Time offset is relative */
#define RESERVE_FLAG_REPLACE SLURM_BIT(18) /* Replace resources
* as assigned to jobs */
#define RESERVE_FLAG_ALL_NODES SLURM_BIT(19) /* Use all compute nodes */
#define RESERVE_FLAG_PURGE_COMP SLURM_BIT(20) /* Purge reservation
* after last job done */
#define RESERVE_FLAG_WEEKDAY SLURM_BIT(21) /* Set WEEKDAY flag */
#define RESERVE_FLAG_NO_WEEKDAY SLURM_BIT(22) /* Clear WEEKDAY flag */
#define RESERVE_FLAG_WEEKEND SLURM_BIT(23) /* Set WEEKEND flag */
#define RESERVE_FLAG_NO_WEEKEND SLURM_BIT(24) /* Clear WEEKEND flag */
#define RESERVE_FLAG_FLEX SLURM_BIT(25) /* Set FLEX flag */
#define RESERVE_FLAG_NO_FLEX SLURM_BIT(26) /* Clear FLEX flag */
#define RESERVE_FLAG_DUR_PLUS SLURM_BIT(27) /* Add duration time,
* only used on
* modifying a
* reservation */
#define RESERVE_FLAG_DUR_MINUS SLURM_BIT(28) /* Remove duration time,
* only used on
* modifying a
* reservation */
#define RESERVE_FLAG_NO_HOLD_JOBS SLURM_BIT(29) /* No hold jobs after
* end of reservation */
#define RESERVE_FLAG_REPLACE_DOWN SLURM_BIT(30) /* Replace DOWN or
* DRAINED nodes */
#define RESERVE_FLAG_NO_PURGE_COMP SLURM_BIT(31) /* Clear PURGE flag */
#define RESERVE_FLAG_MAGNETIC SLURM_BIT(32) /* Allow jobs to run
* without specifying
* the reservation name
* if they meet
* eligibility status */
#define RESERVE_FLAG_NO_MAGNETIC SLURM_BIT(33) /* Clear MAGNETIC flag */
#define RESERVE_FLAG_SKIP SLURM_BIT(34) /* Skip/delete
* next/current
* reservation without
* deleting the
* reservation proper */
#define RESERVE_FLAG_HOURLY SLURM_BIT(35) /* Set HOURLY flag */
#define RESERVE_FLAG_NO_HOURLY SLURM_BIT(36) /* Clear HOURLY flag */
#define RESERVE_FLAG_GRES_REQ SLURM_BIT(37) /* There has been GRES
* requested */
#define RESERVE_TRES_PER_NODE SLURM_BIT(38) /* If the tres_str is per-node
* instead of per-job */
#define RESERVE_FLAG_USER_DEL SLURM_BIT(39) /* Allow users in the ACL to
delete this reservation */
#define RESERVE_FLAG_NO_USER_DEL SLURM_BIT(40) /* Clear USER_DEL flag */
#define RESERVE_FLAG_SCHED_FAILED SLURM_BIT(41) /* Reservation failed to
* schedule job */
#define RESERVE_REOCCURRING (RESERVE_FLAG_HOURLY | RESERVE_FLAG_DAILY | \
RESERVE_FLAG_WEEKLY | RESERVE_FLAG_WEEKDAY | \
RESERVE_FLAG_WEEKEND)
typedef struct resv_core_spec {
char *node_name; /* Name of reserved node */
char *core_id; /* IDs of reserved cores */
} resv_core_spec_t;
typedef struct reserve_info {
char *accounts; /* names of accounts permitted to use */
char *burst_buffer; /* burst buffer resources to be included */
char *comment; /* arbitrary comment */
uint32_t core_cnt; /* count of cores required */
uint32_t core_spec_cnt; /* count of core_spec records */
resv_core_spec_t *core_spec; /* reserved cores specification */
time_t end_time; /* end time of reservation */
char *features; /* required node features */
uint64_t flags; /* see RESERVE_FLAG_* above */
char *groups; /* names of linux Groups permitted to use */
char *licenses; /* names of licenses to be reserved */
uint32_t max_start_delay;/* Maximum delay in which jobs outside of the
* reservation will be permitted to overlap
* once any jobs are queued for the
* reservation */
char *name; /* name of reservation */
uint32_t node_cnt; /* count of nodes required */
int32_t *node_inx; /* list index pairs into node_table for *nodes:
* start_range_1, end_range_1,
* start_range_2, .., -1 */
char *node_list; /* list of reserved nodes or ALL */
char *partition; /* name of partition to be used */
uint32_t purge_comp_time; /* If PURGE_COMP flag is set the amount of
* minutes this reservation will sit idle
* until it is revoked.
*/
time_t start_time; /* start time of reservation */
char *tres_str; /* list of TRES's used by reservation */
char *users; /* names of users permitted to use */
} reserve_info_t;
typedef struct reserve_info_msg {
time_t last_update; /* time of latest info */
uint32_t record_count; /* number of records */
reserve_info_t *reservation_array; /* the reservation records */
} reserve_info_msg_t;
typedef struct resv_desc_msg {
char *accounts; /* names of accounts permitted to use */
char *burst_buffer; /* burst buffer resources to be included */
char *comment; /* arbitrary comment */
uint32_t core_cnt; /* Count of cores required */
uint32_t duration; /* duration of reservation in minutes */
time_t end_time; /* end time of reservation */
char *features; /* required node features */
uint64_t flags; /* see RESERVE_FLAG_* above */
char *groups; /* names of linux groups permitted to use */
void *job_ptr; /* internal use only DON'T PACK */
char *licenses; /* names of licenses to be reserved */
uint32_t max_start_delay;/* Maximum delay in which jobs outside of the
* reservation will be permitted to overlap
* once any jobs are queued for the
* reservation */
char *name; /* name of reservation (optional on create) */
uint32_t node_cnt; /* Count of nodes required. */
char *node_list; /* list of reserved nodes or ALL */
char *partition; /* name of partition to be used */
uint32_t purge_comp_time; /* If PURGE_COMP flag is set the amount of
* minutes this reservation will sit idle
* until it is revoked.
*/
time_t start_time; /* start time of reservation */
char *tres_str; /* list of TRES's used by reservation */
char *users; /* names of users permitted to use */
} resv_desc_msg_t;
typedef struct reserve_response_msg {
char *name; /* name of reservation */
} reserve_response_msg_t;
typedef struct reservation_name_msg {
char *name; /* name of reservation just created or
* to be delete */
} reservation_name_msg_t;
#define DEBUG_FLAG_SELECT_TYPE SLURM_BIT(0) /* SelectType plugin */
#define DEBUG_FLAG_STEPS SLURM_BIT(1) /* slurmctld steps */
#define DEBUG_FLAG_TRIGGERS SLURM_BIT(2) /* slurmctld triggers */
#define DEBUG_FLAG_CPU_BIND SLURM_BIT(3) /* CPU binding */
#define DEBUG_FLAG_NET_RAW SLURM_BIT(4) /* Raw Network dumps */
#define DEBUG_FLAG_NO_CONF_HASH SLURM_BIT(5) /* no warning about
* slurm.conf files checksum
* mismatch */
#define DEBUG_FLAG_GRES SLURM_BIT(6) /* Generic Resource info */
#define DEBUG_FLAG_MPI SLURM_BIT(7) /* MPI debug */
#define DEBUG_FLAG_DATA SLURM_BIT(8) /* data_t logging */
#define DEBUG_FLAG_CONMGR SLURM_BIT(9) /* conmgr logging */
#define DEBUG_FLAG_NET SLURM_BIT(10) /* Network logging */
#define DEBUG_FLAG_PRIO SLURM_BIT(11) /* debug for priority
* plugin */
#define DEBUG_FLAG_BACKFILL SLURM_BIT(12) /* debug for
* sched/backfill */
#define DEBUG_FLAG_GANG SLURM_BIT(13) /* debug gang scheduler */
#define DEBUG_FLAG_RESERVATION SLURM_BIT(14) /* advanced reservations */
#define DEBUG_FLAG_FRONT_END SLURM_BIT(15) /* front-end nodes */
#define DEBUG_FLAG_SACK SLURM_BIT(16) /* SACK handling */
#define DEBUG_FLAG_SWITCH SLURM_BIT(17) /* SwitchType plugin */
#define DEBUG_FLAG_ENERGY SLURM_BIT(18) /* AcctGatherEnergy plugin */
#define DEBUG_FLAG_TLS SLURM_BIT(19) /* TLS plugin */
#define DEBUG_FLAG_LICENSE SLURM_BIT(20) /* AcctGatherProfile
* plugin */
#define DEBUG_FLAG_PROFILE SLURM_BIT(21) /* AcctGatherProfile
* plugin */
#define DEBUG_FLAG_INTERCONNECT SLURM_BIT(22) /* AcctGatherInterconnect
* plugin */
#define DEBUG_FLAG_GLOB_SILENCE SLURM_BIT(23) /* Supresses the glob error */
#define DEBUG_FLAG_JOB_CONT SLURM_BIT(24) /* JobContainer plugin */
#define DEBUG_FLAG_AUDIT_RPCS SLURM_BIT(25) /* Audit RPCs */
#define DEBUG_FLAG_PROTOCOL SLURM_BIT(26) /* Communication protocol */
#define DEBUG_FLAG_BACKFILL_MAP SLURM_BIT(27) /* Backfill scheduler node
* map */
#define DEBUG_FLAG_TRACE_JOBS SLURM_BIT(28) /* Trace jobs by id
* and state */
#define DEBUG_FLAG_ROUTE SLURM_BIT(29) /* Route plugin */
#define DEBUG_FLAG_DB_ASSOC SLURM_BIT(30) /* Association debug */
#define DEBUG_FLAG_DB_EVENT SLURM_BIT(31) /* Event debug */
#define DEBUG_FLAG_DB_JOB SLURM_BIT(32) /* Database job debug */
#define DEBUG_FLAG_DB_QOS SLURM_BIT(33) /* QOS debug */
#define DEBUG_FLAG_DB_QUERY SLURM_BIT(34) /* Database query debug */
#define DEBUG_FLAG_DB_RESV SLURM_BIT(35) /* Reservation debug */
#define DEBUG_FLAG_DB_RES SLURM_BIT(36) /* Resource debug */
#define DEBUG_FLAG_DB_STEP SLURM_BIT(37) /* Database step debug */
#define DEBUG_FLAG_DB_USAGE SLURM_BIT(38) /* Usage/Rollup debug */
#define DEBUG_FLAG_DB_WCKEY SLURM_BIT(39) /* Database WCKey debug */
#define DEBUG_FLAG_BURST_BUF SLURM_BIT(40) /* Burst buffer plugin */
#define DEBUG_FLAG_CPU_FREQ SLURM_BIT(41) /* --cpu_freq debug */
#define DEBUG_FLAG_POWER SLURM_BIT(42) /* Power plugin debug */
/* #define SLURM_BIT(43) /\* UNUSED *\/ */
#define DEBUG_FLAG_DB_ARCHIVE SLURM_BIT(44) /* DBD Archiving/Purging */
#define DEBUG_FLAG_DB_TRES SLURM_BIT(45) /* Database TRES debug */
#define DEBUG_FLAG_JOBCOMP SLURM_BIT(46) /* JobComp debug */
#define DEBUG_FLAG_NODE_FEATURES SLURM_BIT(47) /* Node Features debug */
#define DEBUG_FLAG_FEDR SLURM_BIT(48) /* Federation debug */
#define DEBUG_FLAG_HETJOB SLURM_BIT(49) /* Heterogeneous job debug */
#define DEBUG_FLAG_ACCRUE SLURM_BIT(50) /* Accrue counters debug */
#define DEBUG_FLAG_DBD_AGENT SLURM_BIT(51) /* DBD Agent debug */
#define DEBUG_FLAG_AGENT SLURM_BIT(52) /* RPC Agent debug */
#define DEBUG_FLAG_DEPENDENCY SLURM_BIT(53) /* Dependency debug */
#define DEBUG_FLAG_JAG SLURM_BIT(54) /* Job Account Gather debug */
#define DEBUG_FLAG_CGROUP SLURM_BIT(55) /* cgroup debug */
#define DEBUG_FLAG_SCRIPT SLURM_BIT(56) /* slurmscriptd debug */
#define PREEMPT_MODE_OFF 0x0000 /* disable job preemption */
#define PREEMPT_MODE_SUSPEND 0x0001 /* suspend jobs to preempt */
#define PREEMPT_MODE_REQUEUE 0x0002 /* requeue or kill jobs to preempt */
#define PREEMPT_MODE_CANCEL 0x0008 /* always cancel the job */
#define PREEMPT_MODE_COND_OFF 0x0010 /* represents PREEMPT_MODE_OFF in list*/
#define PREEMPT_MODE_WITHIN 0x4000 /* enable preemption within qos */
#define PREEMPT_MODE_GANG 0x8000 /* enable gang scheduling */
#define RECONFIG_KEEP_PART_INFO SLURM_BIT(0) /* keep dynamic partition info on scontrol reconfig */
#define RECONFIG_KEEP_PART_STAT SLURM_BIT(1) /* keep dynamic partition state on scontrol reconfig */
#define RECONFIG_KEEP_POWER_SAVE_SETTINGS SLURM_BIT(2) /* keep dynamic power save settings on scontrol reconfig */
#define RECONFIG_KEEP_NODE_STATE_FUTURE SLURM_BIT(3) /* Preserve node state on restart/reconfig for FUTURE nodes */
#define HEALTH_CHECK_NODE_IDLE 0x0001 /* execute on idle nodes */
#define HEALTH_CHECK_NODE_ALLOC 0x0002 /* execute on fully allocated nodes */
#define HEALTH_CHECK_NODE_MIXED 0x0004 /* execute on partially allocated nodes */
#define HEALTH_CHECK_NODE_NONDRAINED_IDLE 0x0008 /* execute on idle nodes that
* are not drained */
#define HEALTH_CHECK_CYCLE 0x8000 /* cycle through nodes node */
#define HEALTH_CHECK_NODE_ANY 0x000f /* execute on all node states */
#define PROLOG_FLAG_ALLOC 0x0001 /* execute prolog upon allocation */
#define PROLOG_FLAG_NOHOLD 0x0002 /* don't block salloc/srun until
* slurmctld knows the prolog has
* run on each node in the allocation */
#define PROLOG_FLAG_CONTAIN 0x0004 /* Use proctrack plugin to create a
* container upon allocation */
#define PROLOG_FLAG_SERIAL 0x0008 /* serially execute prolog/epilog */
#define PROLOG_FLAG_X11 0x0010 /* enable slurm x11 forwarding support */
#define PROLOG_FLAG_DEFER_BATCH 0x0020 /* defer REQUEST_BATCH_JOB_LAUNCH until prolog end on all nodes */
#define PROLOG_FLAG_FORCE_REQUEUE_ON_FAIL 0x0040 /* always requeue job on prolog failure */
#define PROLOG_FLAG_RUN_IN_JOB 0x0080 /* run prolog/epilog in slurmstepd */
#define CONF_FLAG_OR SLURM_BIT(0) /* SlurmdParameters=config_overrides */
#define CONF_FLAG_SJC SLURM_BIT(1) /* AccountingStoreFlags=job_comment */
#define CONF_FLAG_DRJ SLURM_BIT(2) /* DisableRootJobs */
#define CONF_FLAG_ASRU SLURM_BIT(3) /* AllowSpecResourcesUsage */
#define CONF_FLAG_PAM SLURM_BIT(4) /* UsePam */
#define CONF_FLAG_WCKEY SLURM_BIT(5) /* TrackWCKey */
#define CONF_FLAG_IPV4_ENABLED SLURM_BIT(6) /* IPv4 is enabled */
#define CONF_FLAG_IPV6_ENABLED SLURM_BIT(7) /* IPv6 is enabled */
#define CONF_FLAG_SJX SLURM_BIT(8) /* AccountingStoreFlags=job_extra */
#define CONF_FLAG_SJS SLURM_BIT(9) /* AccountingStoreFlags=job_script */
#define CONF_FLAG_SJE SLURM_BIT(10) /* AccountingStoreFlags=job_env */
#define CONF_FLAG_L3CSOCK SLURM_BIT(11) /* SlurmdParameters=l3cache_as_socket */
#define CONF_FLAG_NNSOCK SLURM_BIT(12) /* SlurmdParameters=numa_node_as_socket */
#define CONF_FLAG_ECORE SLURM_BIT(13) /* SlurmdParameters=allow_ecores */
#define CONF_FLAG_SHR SLURM_BIT(14) /* SlurmdParameters=shutdown_on_reboot */
#define CONF_FLAG_CONTAIN_SPANK SLURM_BIT(15) /* SlurmdParameters=contain_spank */
#define CONF_FLAG_NO_STDIO SLURM_BIT(16) /* AccountingStoreFlags=no_stdio */
#define LOG_FMT_ISO8601_MS 0
#define LOG_FMT_ISO8601 1
#define LOG_FMT_RFC5424_MS 2
#define LOG_FMT_RFC5424 3
#define LOG_FMT_CLOCK 4
#define LOG_FMT_SHORT 5
#define LOG_FMT_THREAD_ID 6
#define LOG_FMT_RFC3339 7
#define LOG_FMT_FORMAT_STDERR SLURM_BIT(15)
/*
* If adding to slurm_conf_t contents that need to be used in the slurmstepd
* please remember to add those to [un]pack_slurm_conf_lite() in
* src/slurmd/common/slurmstepd_init.[h|c]
*/
typedef struct {
time_t last_update; /* last update time of the build parameters */
char *accounting_storage_tres; /* list of tres */
uint16_t accounting_storage_enforce; /* job requires valid association:
* user/account/partition/cluster */
char *accounting_storage_backup_host; /* accounting storage
* backup host */
char *accounting_storage_ext_host; /* accounting storage ext host */
char *accounting_storage_host; /* accounting storage host */
char *accounting_storage_params; /* accounting storage params */
char *accounting_storage_pass; /* accounting storage
* password */
uint16_t accounting_storage_port;/* node accounting storage port */
char *accounting_storage_type; /* accounting storage type */
char *accounting_storage_user; /* accounting storage user */
void *acct_gather_conf; /* account gather config */
char *acct_gather_energy_type; /* energy accounting type */
char *acct_gather_profile_type; /* profile accounting type */
char *acct_gather_interconnect_type; /* interconnect accounting type */
char *acct_gather_filesystem_type; /* filesystem accounting type */
uint16_t acct_gather_node_freq; /* secs between node acct request */
char *authalttypes; /* alternate authentication types */
char *authinfo; /* authentication info */
char *authalt_params; /* alternate authentication parameters */
char *authtype; /* authentication type */
uint16_t batch_start_timeout; /* max secs for batch job to start */
char *bb_type; /* burst buffer plugin type */
char *bcast_exclude; /* Bcast exclude library paths */
char *bcast_parameters; /* bcast options */
time_t boot_time; /* time slurmctld last booted */
char *certmgr_params; /* certmgr parameters */
char *certmgr_type; /* certmgr type */
void *cgroup_conf; /* cgroup support config file */
char *cli_filter_plugins; /* List of cli_filter plugins to use */
uint16_t cluster_id; /* unique ID for this cluster from the DBD */
char *cluster_name; /* general name of the entire cluster */
char *comm_params; /* Communication parameters */
uint16_t complete_wait; /* seconds to wait for job completion before
* scheduling another job */
uint32_t conf_flags; /* CONF_FLAG_* flags */
char **control_addr; /* comm path of slurmctld
* primary server and backups */
uint32_t control_cnt; /* Length of control_addr & control_machine */
char **control_machine; /* name of slurmctld primary
* server and backups */
uint32_t cpu_freq_def; /* default cpu frequency / governor */
uint32_t cpu_freq_govs; /* cpu freq governors allowed */
char *cred_type; /* credential signature plugin */
char *data_parser_parameters; /* data parser parameters */
uint64_t debug_flags; /* see DEBUG_FLAG_* above for values */
uint64_t def_mem_per_cpu; /* default MB memory per allocated CPU */
char *dependency_params; /* DependencyParameters */
uint16_t eio_timeout; /* timeout for the eio thread */
uint16_t enforce_part_limits; /* if set, reject job exceeding
* partition size and/or time limits */
char **epilog; /* pathname of job epilog run by slurmd */
uint32_t epilog_cnt; /* number of epilog values */
uint32_t epilog_msg_time; /* usecs for slurmctld to process an
* epilog complete message */
char **epilog_slurmctld; /* pathname of job epilog run by slurmctld */
uint32_t epilog_slurmctld_cnt; /* number of epilog_slurmctld values */
char *fed_params; /* Federation parameters */
uint32_t first_job_id; /* first slurm generated job_id to assign */
uint16_t fs_dampening_factor; /* dampening for Fairshare factor */
uint16_t getnameinfo_cache_timeout; /* for getnameinfo() cache*/
uint16_t get_env_timeout; /* timeout for srun --get-user-env option */
char * gres_plugins; /* list of generic resource plugins */
uint16_t group_time; /* update group time interval */
uint16_t group_force; /* update group/partition info even if no change
* detected */
char *gpu_freq_def; /* default GPU frequency / voltage */
char *hash_plugin; /* hash plugin type */
uint32_t hash_val; /* Hash value of the slurm.conf file */
uint16_t health_check_interval; /* secs between health checks */
uint16_t health_check_node_state; /* Node states on which to execute
* health check program, see
* HEALTH_CHECK_NODE_* above */
char * health_check_program; /* pathname of health check program */
uint16_t inactive_limit;/* seconds of inactivity before a
* inactive resource allocation is released */
char *interactive_step_opts; /* InteractiveStepOptions */
char *job_acct_gather_freq; /* poll frequency for job accounting
* gather plugins */
char *job_acct_gather_type; /* job accounting gather type */
char *job_acct_gather_params; /* job accounting gather parameters */
uint16_t job_acct_oom_kill; /* Enforce mem limit at runtime y|n */
char *job_comp_host; /* job completion logging host */
char *job_comp_loc; /* job completion logging location */
char *job_comp_params; /* job completion parameters for plugin */
char *job_comp_pass; /* job completion storage password */
uint32_t job_comp_port; /* job completion storage port */
char *job_comp_type; /* job completion storage type */
char *job_comp_user; /* job completion storage user */
char *job_container_plugin; /* job container plugin type */
list_t *job_defaults_list; /* list of job_defaults_t elements */
uint16_t job_file_append; /* if set, append to stdout/err file */
uint16_t job_requeue; /* If set, jobs get requeued on node failre */
char *job_submit_plugins; /* List of job_submit plugins to use */
uint32_t keepalive_interval; /* Interval between keepalive probes */
uint32_t keepalive_probes; /* Number of keepalive probe attempts */
uint32_t keepalive_time; /* Keep alive time for srun I/O sockets */
uint16_t kill_on_bad_exit; /* If set, the job will be
* terminated immediately when one of
* the processes is aborted or crashed */
uint16_t kill_wait; /* seconds between SIGXCPU to SIGKILL
* on job termination */
char *launch_params; /* step launcher plugin options */
char *licenses; /* licenses available on this cluster */
uint16_t log_fmt; /* Log file timestamp format */
char *mail_domain; /* default domain to append to usernames */
char *mail_prog; /* pathname of mail program */
uint32_t max_array_sz; /* Maximum job array size */
uint32_t max_batch_requeue; /* maximum number of requeues */
uint32_t max_dbd_msgs; /* maximum number of messages queued while DBD
* is not connected */
uint32_t max_job_cnt; /* maximum number of active jobs */
uint32_t max_job_id; /* maximum job id before using first_job_id */
uint64_t max_mem_per_cpu; /* maximum MB memory per allocated CPU */
uint32_t max_node_cnt; /* max number of static + dynamic nodes */
uint32_t max_step_cnt; /* maximum number of steps per job */
uint16_t max_tasks_per_node; /* maximum tasks per node */
char *mcs_plugin; /* mcs plugin type */
char *mcs_plugin_params; /* mcs plugin parameters */
uint32_t min_job_age; /* COMPLETED jobs over this age (secs)
* purged from in memory records */
void *mpi_conf; /* MPI support config file */
char *mpi_default; /* Default version of MPI in use */
char *mpi_params; /* MPI parameters */
uint16_t msg_timeout; /* message timeout */
uint32_t next_job_id; /* next slurm generated job_id to assign */
void *node_features_conf; /* Node Features Plugin config file */
char *node_features_plugins; /* List of node_features plugins to use */
char *node_prefix; /* prefix of nodes in partition, only set in
bluegene clusters NULL otherwise */
uint16_t over_time_limit; /* job's time limit can be exceeded by this
* number of minutes before cancellation */
char *plugindir; /* pathname to plugins */
char *plugstack; /* pathname to plugin stack config file */
uint32_t preempt_exempt_time; /* Time before jobs are preemptable */
uint16_t preempt_mode; /* See PREEMPT_MODE_* in slurm/slurm.h */
char *preempt_params; /* PreemptParameters to tune preemption */
char *preempt_type; /* job preemption selection plugin */
char *prep_params; /* PrEp parameters */
char *prep_plugins; /* PrEp plugins */
uint32_t priority_decay_hl; /* priority decay half life in
* seconds */
uint32_t priority_calc_period; /* seconds between priority decay
* calculation */
uint16_t priority_favor_small; /* favor small jobs over large */
uint16_t priority_flags; /* set some flags for priority configuration,
* see PRIORITY_FLAGS_* above */
uint32_t priority_max_age; /* time when not to add any more
* priority to a job if reached */
char *priority_params; /* priority plugin parameters */
uint16_t priority_reset_period; /* when to clear usage,
* see PRIORITY_RESET_* */
char *priority_type; /* priority type plugin */
uint32_t priority_weight_age; /* weight for age factor */
uint32_t priority_weight_assoc; /* weight for assoc factor */
uint32_t priority_weight_fs; /* weight for Fairshare factor */
uint32_t priority_weight_js; /* weight for Job Size factor */
uint32_t priority_weight_part; /* weight for Partition factor */
uint32_t priority_weight_qos; /* weight for QOS factor */
char *priority_weight_tres; /* weights (str) for different TRES' */
uint16_t private_data; /* block viewing of information,
* see PRIVATE_DATA_* */
char *proctrack_type; /* process tracking plugin type */
char **prolog; /* pathname of job prolog run by slurmd */
uint32_t prolog_cnt; /* number of prolog values */
uint16_t prolog_epilog_timeout; /* prolog/epilog timeout */
char **prolog_slurmctld; /* pathname of job prolog run by slurmctld */
uint32_t prolog_slurmctld_cnt; /* number of prolog_slurmctld values */
uint16_t propagate_prio_process; /* process priority propagation,
* see PROP_PRIO_* */
uint16_t prolog_flags; /* set some flags for prolog configuration
see PROLOG_FLAG_* */
char *propagate_rlimits;/* Propagate (all/specific) resource limits */
char *propagate_rlimits_except;/* Propagate all rlimits except these */
char *reboot_program; /* program to reboot the node */
uint16_t reconfig_flags;/* see RECONFIG_* */
char *requeue_exit; /* requeue exit values */
char *requeue_exit_hold; /* requeue exit hold values */
char *resume_fail_program; /* program to handle failed resume tries */
char *resume_program; /* program to make nodes full power */
uint16_t resume_rate; /* nodes to make full power, per minute */
uint16_t resume_timeout;/* time required in order to perform a node
* resume operation */
char *resv_epilog; /* path of reservation epilog run by slurmctld */
uint16_t resv_over_run; /* how long a running job can exceed
* reservation time */
char *resv_prolog; /* path of reservation prolog run by slurmctld */
uint16_t ret2service; /* 1 return DOWN node to service at
* registration */
char *sched_logfile; /* where slurm Scheduler log gets written */
uint16_t sched_log_level; /* configured level of slurm Scheduler log */
char *sched_params; /* SchedulerParameters OR
* contents of scheduler plugin config file */
uint16_t sched_time_slice; /* gang scheduler slice time, secs */
char *schedtype; /* type of scheduler to use */
char *scron_params; /* ScronParameters */
char *select_type; /* type of node selector to use */
void *select_conf_key_pairs; /* key-pair list which can be
* listed with slurm_print_key_pairs() */
uint16_t select_type_param; /* Parameters
* describing the select_type plugin */
char *site_factor_plugin; /* PrioritySiteFactorPlugin */
char *site_factor_params; /* PrioritySiteFactorParameters */
char *slurm_conf; /* pathname of slurm config file */
uint32_t slurm_user_id; /* uid of slurm_user_name */
char *slurm_user_name; /* user that slurmctld runs as */
uint32_t slurmd_user_id;/* uid of slurmd_user_name */
char *slurmd_user_name; /* user that slurmd runs as */
char *slurmctld_addr; /* Address used for communications to the
* currently active slurmctld daemon */
uint16_t slurmctld_debug; /* slurmctld logging level */
char *slurmctld_logfile;/* where slurmctld error log gets written */
char *slurmctld_pidfile;/* where to put slurmctld pidfile */
uint32_t slurmctld_port; /* default communications port to slurmctld */
uint16_t slurmctld_port_count; /* number of slurmctld comm ports */
char *slurmctld_primary_off_prog; /* Run when becomes slurmctld backup */
char *slurmctld_primary_on_prog; /* Run when becomes slurmctld primary */
uint16_t slurmctld_syslog_debug; /* slurmctld output to
* local logfile and syslog*/
uint16_t slurmctld_timeout;/* seconds that backup controller waits
* on non-responding primarly controller */
char *slurmctld_params; /* SlurmctldParameters */
uint16_t slurmd_debug; /* slurmd logging level */
char *slurmd_logfile; /* where slurmd error log gets written */
char *slurmd_params; /* SlurmdParameters */
char *slurmd_pidfile; /* where to put slurmd pidfile */
uint32_t slurmd_port; /* default communications port to slurmd */
char *slurmd_spooldir; /* where slurmd put temporary state info */
uint16_t slurmd_syslog_debug; /* slurmd output to
* local logfile and syslog*/
uint16_t slurmd_timeout;/* how long slurmctld waits for slurmd before
* considering node DOWN */
char *srun_epilog; /* srun epilog program */
uint16_t *srun_port_range; /* port range for srun */
char *srun_prolog; /* srun prolog program */
char *state_save_location;/* pathname of slurmctld state save
* directory */
char *suspend_exc_nodes;/* nodes to not make power saving */
char *suspend_exc_parts;/* partitions to not make power saving */
char *suspend_exc_states; /* states that should not be powered down */
char *suspend_program; /* program to make nodes power saving */
uint16_t suspend_rate; /* nodes to make power saving, per minute */
uint32_t suspend_time; /* node idle for this long before power save mode */
uint16_t suspend_timeout;/* time required in order to perform a node
* suspend operation */
char *switch_type; /* switch or interconnect type */
char *switch_param; /* SwitchParameters */
char *task_epilog; /* pathname of task launch epilog */
char *task_plugin; /* task launch plugin */
uint32_t task_plugin_param; /* see CPU_BIND_* */
char *task_prolog; /* pathname of task launch prolog */
char *tls_params; /* TLSParameters */
char *tls_type; /* tls plugin type */
uint16_t tcp_timeout; /* tcp timeout */
char *tmp_fs; /* pathname of temporary file system */
char *topology_param; /* network topology parameters */
char *topology_plugin; /* network topology plugin */
uint16_t tree_width; /* number of threads per node to span */
char *unkillable_program; /* program run by the slurmstepd when
* processes in a job step are unkillable */
uint16_t unkillable_timeout; /* time in seconds, after processes in a
* job step have been signaled, before
* they are considered "unkillable". */
char *version; /* version of slurmctld */
uint16_t vsize_factor; /* virtual memory limit size factor */
uint16_t wait_time; /* default job --wait time */
char *x11_params; /* X11Parameters */
} slurm_conf_t;
typedef struct slurmd_status_msg {
time_t booted; /* when daemon was started */
time_t last_slurmctld_msg; /* time of last slurmctld message */
uint16_t slurmd_debug; /* logging level */
uint16_t actual_cpus; /* actual logical processor count */
uint16_t actual_boards; /* actual total boards count */
uint16_t actual_sockets; /* actual total sockets count */
uint16_t actual_cores; /* actual core per socket count */
uint16_t actual_threads; /* actual thread per core count */
uint64_t actual_real_mem; /* actual real memory in MB */
uint32_t actual_tmp_disk; /* actual temp disk space in MB */
uint32_t pid; /* process ID */
char *hostname; /* local hostname */
char *slurmd_logfile; /* slurmd log file location */
char *step_list; /* list of active job steps */
char *version; /* version running */
} slurmd_status_t;
typedef struct submit_response_msg {
uint32_t job_id; /* job ID */
uint32_t step_id; /* step ID */
uint32_t error_code; /* error code for warning message */
char *job_submit_user_msg; /* job submit plugin user_msg */
} submit_response_msg_t;
/* NOTE: If setting node_addr and/or node_hostname then comma separate names
* and include an equal number of node_names */
typedef struct slurm_update_node_msg {
char *comment; /* arbitrary comment */
uint32_t cpu_bind; /* default CPU binding type */
char *cert_token; /* node's unique token for certmgr validation */
char *extra; /* arbitrary string */
char *features; /* new available feature for node */
char *features_act; /* new active feature for node */
char *gres; /* new generic resources for node */
char *instance_id; /* cloud instance id */
char *instance_type; /* cloud instance type */
char *node_addr; /* communication name (optional) */
char *node_hostname; /* node's hostname (optional) */
char *node_names; /* nodelist expression */
uint32_t node_state; /* see enum node_states */
char *reason; /* reason for node being DOWN or DRAINING */
uint32_t reason_uid; /* user ID of sending (needed if user
* root is sending message) */
uint32_t resume_after; /* automatically resume DOWN or DRAINED node
* after this amount of seconds */
uint32_t weight; /* new weight for node */
} update_node_msg_t;
typedef struct slurm_update_front_end_msg {
char *name; /* comma separated list of front end nodes */
uint32_t node_state; /* see enum node_states */
char *reason; /* reason for node being DOWN or DRAINING */
uint32_t reason_uid; /* user ID of sending (needed if user
* root is sending message) */
} update_front_end_msg_t;
typedef struct partition_info update_part_msg_t;
typedef struct job_sbcast_cred_msg {
uint32_t job_id; /* assigned job id */
char *node_list; /* assigned list of nodes */
void *sbcast_cred; /* opaque data structure */
} job_sbcast_cred_msg_t;
typedef struct sbcast_cred_req_msg {
char *node_list;
} sbcast_cred_req_msg_t;
typedef struct {
uint32_t lifespan;
char *username;
} token_request_msg_t;
typedef struct {
char *token;
} token_response_msg_t;
/* Opaque data type for slurm_step_ctx_* functions */
typedef struct slurm_step_ctx_struct slurm_step_ctx_t;
#define STAT_COMMAND_RESET 0x0000
#define STAT_COMMAND_GET 0x0001
typedef struct stats_info_request_msg {
uint16_t command_id;
} stats_info_request_msg_t;
typedef struct stats_info_response_msg {
uint32_t parts_packed;
time_t req_time;
time_t req_time_start;
uint32_t server_thread_count;
uint32_t agent_queue_size;
uint32_t agent_count;
uint32_t agent_thread_count;
uint32_t dbd_agent_queue_size;
uint32_t gettimeofday_latency;
uint32_t schedule_cycle_max;
uint32_t schedule_cycle_last;
uint32_t schedule_cycle_sum;
uint32_t schedule_cycle_counter;
uint32_t schedule_cycle_depth;
uint32_t *schedule_exit;
uint32_t schedule_exit_cnt;
uint32_t schedule_queue_len;
uint32_t jobs_submitted;
uint32_t jobs_started;
uint32_t jobs_completed;
uint32_t jobs_canceled;
uint32_t jobs_failed;
uint32_t jobs_pending;
uint32_t jobs_running;
time_t job_states_ts;
uint32_t bf_backfilled_jobs;
uint32_t bf_last_backfilled_jobs;
uint32_t bf_backfilled_het_jobs;
uint32_t bf_cycle_counter;
uint64_t bf_cycle_sum;
uint32_t bf_cycle_last;
uint32_t bf_cycle_max;
uint32_t *bf_exit;
uint32_t bf_exit_cnt;
uint32_t bf_last_depth;
uint32_t bf_last_depth_try;
uint32_t bf_depth_sum;
uint32_t bf_depth_try_sum;
uint32_t bf_queue_len;
uint32_t bf_queue_len_sum;
uint32_t bf_table_size;
uint32_t bf_table_size_sum;
time_t bf_when_last_cycle;
uint32_t bf_active;
uint32_t rpc_type_size;
uint16_t *rpc_type_id;
uint32_t *rpc_type_cnt;
uint64_t *rpc_type_time;
uint8_t rpc_queue_enabled;
uint16_t *rpc_type_queued;
uint64_t *rpc_type_dropped;
uint16_t *rpc_type_cycle_last;
uint16_t *rpc_type_cycle_max;
uint32_t rpc_user_size;
uint32_t *rpc_user_id;
uint32_t *rpc_user_cnt;
uint64_t *rpc_user_time;
uint32_t rpc_queue_type_count;
uint32_t *rpc_queue_type_id;
uint32_t *rpc_queue_count;
uint32_t rpc_dump_count;
uint32_t *rpc_dump_types;
char **rpc_dump_hostlist;
} stats_info_response_msg_t;
#define TRIGGER_FLAG_PERM 0x0001
#define TRIGGER_RES_TYPE_JOB 0x0001
#define TRIGGER_RES_TYPE_NODE 0x0002
#define TRIGGER_RES_TYPE_SLURMCTLD 0x0003
#define TRIGGER_RES_TYPE_SLURMDBD 0x0004
#define TRIGGER_RES_TYPE_DATABASE 0x0005
#define TRIGGER_RES_TYPE_FRONT_END 0x0006
#define TRIGGER_RES_TYPE_OTHER 0x0007
#define TRIGGER_TYPE_UP SLURM_BIT(0)
#define TRIGGER_TYPE_DOWN SLURM_BIT(1)
#define TRIGGER_TYPE_FAIL SLURM_BIT(2)
#define TRIGGER_TYPE_TIME SLURM_BIT(3)
#define TRIGGER_TYPE_FINI SLURM_BIT(4)
#define TRIGGER_TYPE_RECONFIG SLURM_BIT(5)
/* SLURM_BIT(6), UNUSED */
#define TRIGGER_TYPE_IDLE SLURM_BIT(7)
#define TRIGGER_TYPE_DRAINED SLURM_BIT(8)
#define TRIGGER_TYPE_PRI_CTLD_FAIL SLURM_BIT(9)
#define TRIGGER_TYPE_PRI_CTLD_RES_OP SLURM_BIT(10)
#define TRIGGER_TYPE_PRI_CTLD_RES_CTRL SLURM_BIT(11)
#define TRIGGER_TYPE_PRI_CTLD_ACCT_FULL SLURM_BIT(12)
#define TRIGGER_TYPE_BU_CTLD_FAIL SLURM_BIT(13)
#define TRIGGER_TYPE_BU_CTLD_RES_OP SLURM_BIT(14)
#define TRIGGER_TYPE_BU_CTLD_AS_CTRL SLURM_BIT(15)
#define TRIGGER_TYPE_PRI_DBD_FAIL SLURM_BIT(16)
#define TRIGGER_TYPE_PRI_DBD_RES_OP SLURM_BIT(17)
#define TRIGGER_TYPE_PRI_DB_FAIL SLURM_BIT(18)
#define TRIGGER_TYPE_PRI_DB_RES_OP SLURM_BIT(19)
#define TRIGGER_TYPE_BURST_BUFFER SLURM_BIT(20)
#define TRIGGER_TYPE_DRAINING SLURM_BIT(21)
#define TRIGGER_TYPE_RESUME SLURM_BIT(22)
typedef struct trigger_info {
uint16_t flags; /* TRIGGER_FLAG_* */
uint32_t trig_id; /* trigger ID */
uint16_t res_type; /* TRIGGER_RES_TYPE_* */
char * res_id; /* resource ID */
uint32_t control_inx; /* controller index */
uint32_t trig_type; /* TRIGGER_TYPE_* */
uint16_t offset; /* seconds from trigger, 0x8000 origin */
uint32_t user_id; /* user requesting trigger */
char * program; /* program to execute */
} trigger_info_t;
typedef struct trigger_info_msg {
uint32_t record_count; /* number of records */
trigger_info_t *trigger_array; /* the trigger records */
} trigger_info_msg_t;
/* Individual license information
*/
typedef struct slurm_license_info {
char *name; /* license name */
uint32_t total; /* total number of available licenses */
uint32_t in_use; /* number of license in use */
uint32_t available; /* number of available license */
uint8_t remote; /* non-zero if remote license (not
* defined in slurm.conf) */
uint32_t reserved; /* number of licenses reserved */
uint32_t last_consumed; /* number of licenses last known to be
consumed in the license manager
(for remote) */
uint32_t last_deficit;
time_t last_update; /* last updated (for remote) */
} slurm_license_info_t;
/* License information array as returned by the controller.
*/
typedef struct license_info_msg {
time_t last_update;
uint32_t num_lic;
slurm_license_info_t *lic_array;
} license_info_msg_t;
typedef struct {
uint32_t job_array_count;
char **job_array_id; /* Note: The string may be truncated */
uint32_t *error_code;
char **err_msg;
} job_array_resp_msg_t;
/* Association manager state running in the slurmctld */
typedef struct {
list_t *assoc_list; /* list of slurmdb_assoc_rec_t with usage packed */
list_t *qos_list; /* list of slurmdb_qos_rec_t with usage packed */
uint32_t tres_cnt;
char **tres_names;
list_t *user_list; /* list of slurmdb_user_rec_t */
} assoc_mgr_info_msg_t;
#define ASSOC_MGR_INFO_FLAG_ASSOC 0x00000001
#define ASSOC_MGR_INFO_FLAG_USERS 0x00000002
#define ASSOC_MGR_INFO_FLAG_QOS 0x00000004
typedef struct {
list_t *acct_list; /* char * list of account names */
uint32_t flags; /* flags determining what is returned */
list_t *qos_list; /* char * list of qos names */
list_t *user_list; /* char * list of user names */
} assoc_mgr_info_request_msg_t;
typedef struct network_callerid_msg {
unsigned char ip_src[16];
unsigned char ip_dst[16];
uint32_t port_src;
uint32_t port_dst;
int32_t af; /* NOTE: un/packed as uint32_t */
} network_callerid_msg_t;
/*****************************************************************************\
* SLURM CLUSTER FUNCTIONS
\*****************************************************************************/
/*
* slurm_get_cluster_info - get list of cluster records
* OUT cluster_records - list of slurmdb_cluster_rec_t
* IN cluster_names - char list of cluster names. Set to NULL for all clusters
* IN show_flags - set SHOW_FEDERATION to get cluster info from slurmctld when
* running in a federated setup. Otherwise, cluster info is retrieved from
* the slurmdbd.
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR
* NOTE: free the response using slurm_free_node_info_msg
*/
extern int slurm_get_cluster_info(list_t **cluster_records, char *cluster_names,
uint16_t show_flags);
/*****************************************************************************\
* RESOURCE ALLOCATION FUNCTIONS
\*****************************************************************************/
/*
* slurm_init_job_desc_msg - initialize job descriptor with
* default values
* OUT job_desc_msg - user defined job descriptor
*/
extern void slurm_init_job_desc_msg(job_desc_msg_t *job_desc_msg);
/*
* slurm_allocate_resources - allocate resources for a job request
* If the requested resources are not immediately available, the slurmctld
* will send the job_alloc_resp_msg to the specified node and port.
* IN job_desc_msg - description of resource allocation request
* OUT job_alloc_resp_msg - response to request. This only represents
* a job allocation if resources are immediately. Otherwise it just contains
* the job id of the enqueued job request.
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
* NOTE: free the response using slurm_free_resource_allocation_response_msg()
*/
extern int slurm_allocate_resources(job_desc_msg_t *job_desc_msg,
resource_allocation_response_msg_t **job_alloc_resp_msg);
/*
* slurm_allocate_resources_blocking
* allocate resources for a job request. This call will block until
* the allocation is granted, or the specified timeout limit is reached.
* IN req - description of resource allocation request
* IN timeout - amount of time, in seconds, to wait for a response before
* giving up.
* A timeout of zero will wait indefinitely.
* IN pending_callback - If the allocation cannot be granted immediately,
* the controller will put the job in the PENDING state. If
* pending callback is not NULL, it will be called with the job_id
* of the pending job as the sole parameter.
*
* RET allocation structure on success, NULL on error set errno to
* indicate the error (errno will be ETIMEDOUT if the timeout is reached
* with no allocation granted)
* NOTE: free the response using slurm_free_resource_allocation_response_msg()
*/
extern resource_allocation_response_msg_t *slurm_allocate_resources_blocking(
const job_desc_msg_t *user_req,
time_t timeout,
void (*pending_callback)(uint32_t job_id));
/*
* slurm_free_resource_allocation_response_msg - free slurm resource
* allocation response message
* IN msg - pointer to allocation response message
* NOTE: buffer is loaded by slurm_allocate_resources
*/
extern void slurm_free_resource_allocation_response_msg(resource_allocation_response_msg_t *msg);
/*
* slurm_allocate_het_job_blocking
* allocate resources for a list of job requests. This call will block
* until the entire allocation is granted, or the specified timeout limit
* is reached.
* IN job_req_list - list of resource allocation requests, type job_desc_msg_t
* IN timeout - amount of time, in seconds, to wait for a response before
* giving up.
* A timeout of zero will wait indefinitely.
* IN pending_callback - If the allocation cannot be granted immediately,
* the controller will put the job in the PENDING state. If
* pending callback is not NULL, it will be called with the job_id
* of the pending job as the sole parameter.
*
* RET list of allocation structures on success, NULL on error set errno to
* indicate the error (errno will be ETIMEDOUT if the timeout is reached
* with no allocation granted)
* NOTE: free the response using list_destroy()
*/
extern list_t *slurm_allocate_het_job_blocking(
list_t *job_req_list,
time_t timeout,
void(*pending_callback)(uint32_t job_id));
/*
* slurm_allocation_lookup - retrieve info for an existing resource
* allocation
* IN job_id - job allocation identifier
* OUT resp - job allocation information
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
* NOTE: free the response using slurm_free_resource_allocation_response_msg()
*/
extern int slurm_allocation_lookup(uint32_t job_id,
resource_allocation_response_msg_t **resp);
/*
* slurm_het_job_lookup - retrieve info for an existing heterogeneous job
* allocation without the addrs and such
* IN jobid - job allocation identifier
* OUT resp - list of job allocation information, type
* resource_allocation_response_msg_t
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
* NOTE: returns information an individual job as well
* NOTE: free the response using list_destroy()
*/
extern int slurm_het_job_lookup(uint32_t jobid, list_t **resp);
/*
* slurm_read_hostfile - Read a Slurm hostfile specified by "filename".
* "filename" must contain a list of Slurm NodeNames, one per line.
* Reads up to "n" number of hostnames from the file. Returns a
* string representing a hostlist ranged string of the contents of
* the file. This is a helper function, it does not contact any
* Slurm daemons.
*
* IN filename - name of Slurm Hostlist file to be read.
* IN n - number of NodeNames required
* RET - a string representing the hostlist. Returns NULL if there are
* fewer than "n" hostnames in the file, or if an error occurs.
*
* NOTE: Returned string must be freed with free().
*/
extern char *slurm_read_hostfile(const char *filename, int n);
/*
* slurm_allocation_msg_thr_create - startup a message handler talking
* with the controller dealing with messages from the controller during an
* allocation.
* IN port - port we are listening for messages on from the controller
* IN callbacks - callbacks for different types of messages
* RET allocation_msg_thread_t * or NULL on failure
*/
extern allocation_msg_thread_t *slurm_allocation_msg_thr_create(uint16_t *port,
const slurm_allocation_callbacks_t *callbacks);
/*
* slurm_allocation_msg_thr_destroy - shutdown the message handler talking
* with the controller dealing with messages from the controller during an
* allocation.
* IN msg_thr - allocation_msg_thread_t pointer allocated with
* slurm_allocation_msg_thr_create
*/
extern void slurm_allocation_msg_thr_destroy(allocation_msg_thread_t *msg_thr);
/*
* slurm_submit_batch_job - issue RPC to submit a job for later execution
* NOTE: free the response using slurm_free_submit_response_response_msg
* IN job_desc_msg - description of batch job request
* OUT slurm_alloc_msg - response to request
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_submit_batch_job(job_desc_msg_t *job_desc_msg,
submit_response_msg_t **slurm_alloc_msg);
/*
* slurm_submit_batch_het_job - issue RPC to submit a heterogeneous job for
* later execution
* NOTE: free the response using slurm_free_submit_response_response_msg
* IN job_req_list - list of resource allocation requests, type job_desc_msg_t
* OUT slurm_alloc_msg - response to request
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_submit_batch_het_job(list_t *job_req_list,
submit_response_msg_t **slurm_alloc_msg);
/*
* slurm_free_submit_response_response_msg - free slurm
* job submit response message
* IN msg - pointer to job submit response message
* NOTE: buffer is loaded by slurm_submit_batch_job
*/
extern void slurm_free_submit_response_response_msg(submit_response_msg_t *msg);
/*
* slurm_job_batch_script - retrieve the batch script for a given jobid
* returns SLURM_SUCCESS, or appropriate error code
*/
extern int slurm_job_batch_script(FILE *out, uint32_t jobid);
/*
* slurm_job_will_run - determine if a job would execute immediately if
* submitted now
* IN job_desc_msg - description of resource allocation request
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_job_will_run(job_desc_msg_t *job_desc_msg);
/*
* slurm_het_job_will_run - determine if a heterogeneous job would execute
* immediately if submitted now
* IN job_req_list - list of job_desc_msg_t structures describing the resource
* allocation request
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_het_job_will_run(list_t *job_req_list);
/*
* slurm_job_will_run2 - determine if a job would execute immediately if
* submitted now
* IN job_desc_msg - description of resource allocation request
* OUT will_run_resp - job run time data
* free using slurm_free_will_run_response_msg()
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_job_will_run2(job_desc_msg_t *req,
will_run_response_msg_t **will_run_resp);
/*
* slurm_sbcast_lookup - retrieve info for an existing resource allocation
* including a credential needed for sbcast.
* IN selected_step - filled in with step_id and het_job_offset
* OUT info - job allocation information including a credential for sbcast
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
* NOTE: free the "resp" using slurm_free_sbcast_cred_msg
*/
extern int slurm_sbcast_lookup(slurm_selected_step_t *selected_step,
job_sbcast_cred_msg_t **info);
extern void slurm_free_sbcast_cred_msg(job_sbcast_cred_msg_t *msg);
/* slurm_load_licenses()
*
* Retrieve license information from the controller.
* IN feature - feature name or NULL
* OUT
*
*/
extern int slurm_load_licenses(time_t, license_info_msg_t **, uint16_t);
extern void slurm_free_license_info_msg(license_info_msg_t *);
/* get the running assoc_mgr info
* IN assoc_mgr_info_request_msg_t: request filtering data returned
* OUT assoc_mgr_info_msg_t: returned structure filled in with
* assoc_mgr lists, must be freed by slurm_free_assoc_mgr_info_msg
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_load_assoc_mgr_info(assoc_mgr_info_request_msg_t *,
assoc_mgr_info_msg_t **);
extern void slurm_free_assoc_mgr_info_msg(assoc_mgr_info_msg_t *);
extern void slurm_free_assoc_mgr_info_request_members(assoc_mgr_info_request_msg_t *);
extern void slurm_free_assoc_mgr_info_request_msg(assoc_mgr_info_request_msg_t *);
/*****************************************************************************\
* JOB/STEP SIGNALING FUNCTIONS
\*****************************************************************************/
typedef struct job_step_kill_msg {
char *sjob_id;
uint16_t signal;
uint16_t flags; /* KILL_* flags below */
char *sibling;
slurm_step_id_t step_id;
} job_step_kill_msg_t;
typedef struct {
char *account;
uint16_t flags; /* KILL_* flags below */
char *job_name;
char **jobs_array;
uint32_t jobs_cnt;
char *partition;
char *qos;
char *reservation;
uint16_t signal;
uint32_t state;
uint32_t user_id;
char *user_name;
char *wckey;
char *nodelist;
} kill_jobs_msg_t;
#define KILL_JOB_MSG_INITIALIZER \
{ \
.signal = SIGKILL, \
.state = JOB_END, \
.user_id = SLURM_AUTH_NOBODY, \
}
typedef struct {
uint32_t error_code;
char *error_msg;
slurm_selected_step_t *id; /* Note: if the job is an array job, then
* id.step_id.job_id is the array_job_id or
* het_job_id, respectively. */
uint32_t real_job_id; /* The actual job id if the job was found, or
* NO_VAL if the job was not found */
char *sibling_name;
} kill_jobs_resp_job_t;
typedef struct {
kill_jobs_resp_job_t *job_responses;
uint32_t jobs_cnt;
} kill_jobs_resp_msg_t;
/*
* NOTE: See _signal_batch_job() controller and _rpc_signal_tasks() in slurmd.
*/
#define KILL_JOB_BATCH SLURM_BIT(0) /* signal batch shell only */
#define KILL_ARRAY_TASK SLURM_BIT(1) /* kill single task of a job array */
#define KILL_STEPS_ONLY SLURM_BIT(2) /* Do not signal batch script */
#define KILL_FULL_JOB SLURM_BIT(3) /* Signal all steps, including batch
* script */
#define KILL_FED_REQUEUE SLURM_BIT(4) /* Mark job as requeued when requeued */
#define KILL_HURRY SLURM_BIT(5) /* Skip burst buffer stage out */
#define KILL_OOM SLURM_BIT(6) /* Kill due to Out-Of-Memory */
#define KILL_NO_SIBS SLURM_BIT(7) /* Don't kill other sibling jobs */
#define KILL_JOB_RESV SLURM_BIT(8) /* Job is willing to run on nodes in a
* magnetic reservation. */
#define KILL_NO_CRON SLURM_BIT(9) /* Do not request killing cron Jobs.
* Remove this two versions after 23.11.
*/
#define KILL_NO_SIG_FAIL SLURM_BIT(10) /* Don't fail job due to signal (steps only) */
#define KILL_JOBS_VERBOSE SLURM_BIT(11) /* Verbose response requested */
#define KILL_CRON SLURM_BIT(12) /* Request killing cron jobs */
/* Use top bit of uint16_t in conjuction with KILL_* flags to indicate signal
* has been sent to job previously. Does not need to be passed to slurmd. */
#define WARN_SENT SLURM_BIT(15) /* warn already sent, clear this on
* requeue */
/*
* slurm_kill_job - send the specified signal to all steps of an existing job
* IN job_id - the job's id
* IN signal - signal number
* IN flags - see KILL_JOB_* flags above
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_kill_job(uint32_t job_id, uint16_t signal, uint16_t flags);
/*
* slurm_kill_job_step - send the specified signal to an existing job step
* IN job_id - the job's id
* IN step_id - the job step's id
* IN signal - signal number
* IN flags - see KILL_* or 0 for no flags
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_kill_job_step(uint32_t job_id, uint32_t step_id,
uint16_t signal, uint16_t flags);
/*
* slurm_kill_job2 - send REQUEST_KILL_JOB msg to an existing job or step.
* IN job_id - the job's id (in a string format)
* IN signal - signal to send
* IN flags - see KILL_* flags above (such as KILL_JOB_BATCH)
* IN sibling - optional string of sibling cluster to send the message to.
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_kill_job2(const char *job_id, uint16_t signal, uint16_t flags,
const char *sibling);
/*
* slurm_kill_jobs - Kill all jobs matching the filters specified in msg.
* IN kill_msg - Message that specifies the jobs to be killed.
* IN/OUT kill_msg_resp - Pointer to a response message, filled in by the api.
* RET SLURM_SUCCESS on success, otherwise return an error.
*/
extern int slurm_kill_jobs(kill_jobs_msg_t *kill_msg,
kill_jobs_resp_msg_t **kill_msg_resp);
/*
* slurm_signal_job - send the specified signal to all steps of an existing job
* IN job_id - the job's id
* IN signal - signal number
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_signal_job(uint32_t job_id, uint16_t signal);
/*
* slurm_signal_job_step - send the specified signal to an existing job step
* IN job_id - the job's id
* IN step_id - the job step's id - use SLURM_BATCH_SCRIPT as the step_id
* to send a signal to a job's batch script
* IN signal - signal number
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_signal_job_step(uint32_t job_id,
uint32_t step_id,
uint32_t signal);
/*****************************************************************************\
* JOB/STEP COMPLETION FUNCTIONS
\*****************************************************************************/
/*
* slurm_complete_job - note the completion of a job and all of its steps
* IN job_id - the job's id
* IN job_return_code - the highest exit code of any task of the job
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_complete_job(uint32_t job_id, uint32_t job_return_code);
/*
* slurm_terminate_job_step - terminates a job step by sending a
* REQUEST_TERMINATE_TASKS rpc to all slurmd of a job step, and then
* calls slurm_complete_job_step() after verifying that all
* nodes in the job step no longer have running tasks from the job
* step. (May take over 35 seconds to return.)
* IN job_id - the job's id
* IN step_id - the job step's id - use SLURM_BATCH_SCRIPT as the step_id
* to terminate a job's batch script
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_terminate_job_step(uint32_t job_id, uint32_t step_id);
/*****************************************************************************\
* SLURM TASK SPAWNING FUNCTIONS
\*****************************************************************************/
/*
* slurm_step_launch_params_t_init - initialize a user-allocated
* slurm_step_launch_params_t structure with default values.
* default values. This function will NOT allocate any new memory.
* IN ptr - pointer to a structure allocated by the use. The structure will
* be initialized.
*/
extern void slurm_step_launch_params_t_init(slurm_step_launch_params_t *ptr);
/*
* slurm_step_launch - launch a parallel job step
* IN ctx - job step context generated by slurm_step_ctx_create
* IN params - job step parameters
* IN callbacks - Identify functions to be called when various events occur
* IN het_job_step_cnt - Total count of hetjob steps to be launched, -1 otherwise
* RET SLURM_SUCCESS or SLURM_ERROR (with errno set)
*/
extern int slurm_step_launch(slurm_step_ctx_t *ctx,
const slurm_step_launch_params_t *params,
const slurm_step_launch_callbacks_t *callbacks);
/*
* slurm_step_launch_add - Add tasks to a step that was already started
* IN ctx - job step context generated by slurm_step_ctx_create
* IN first_ctx - job step context generated by slurm_step_ctx_create for
* first component of the job step
* IN params - job step parameters
* IN node_list - list of extra nodes to add
* IN start_nodeid - in the global scheme which node id is the first
* node in node_list.
* RET SLURM_SUCCESS or SLURM_ERROR (with errno set)
*/
extern int slurm_step_launch_add(slurm_step_ctx_t *ctx,
slurm_step_ctx_t *first_ctx,
const slurm_step_launch_params_t *params,
char *node_list);
/*
* Block until all tasks have started.
*/
extern int slurm_step_launch_wait_start(slurm_step_ctx_t *ctx);
/*
* Block until all tasks have finished (or failed to start altogether).
*/
extern void slurm_step_launch_wait_finish(slurm_step_ctx_t *ctx);
/*
* Abort an in-progress launch, or terminate the fully launched job step.
*
* Can be called from a signal handler.
*/
extern void slurm_step_launch_abort(slurm_step_ctx_t *ctx);
/*
* Forward a signal to all those nodes with running tasks
*/
extern void slurm_step_launch_fwd_signal(slurm_step_ctx_t *ctx, int signo);
/*
* Wake tasks stopped for debugging on nodes with running tasks
*/
extern void slurm_step_launch_fwd_wake(slurm_step_ctx_t *ctx);
/*****************************************************************************\
* SLURM CONTROL CONFIGURATION READ/PRINT/UPDATE FUNCTIONS
\*****************************************************************************/
/*
* slurm_api_version - Return a single number reflecting the Slurm API's
* version number. Use the macros SLURM_VERSION_NUM, SLURM_VERSION_MAJOR,
* SLURM_VERSION_MINOR, and SLURM_VERSION_MICRO to work with this value
* RET API's version number
*/
extern long slurm_api_version(void);
/*
* slurm_load_ctl_conf - issue RPC to get slurm control configuration
* information if changed since update_time
* IN update_time - time of current configuration data
* IN slurm_ctl_conf_ptr - place to store slurm control configuration
* pointer
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
* NOTE: free the response using slurm_free_ctl_conf
*/
extern int slurm_load_ctl_conf(time_t update_time,
slurm_conf_t **slurm_ctl_conf_ptr);
/*
* slurm_free_ctl_conf - free slurm control information response message
* IN msg - pointer to slurm control information response message
* NOTE: buffer is loaded by slurm_load_ctl_conf
*/
extern void slurm_free_ctl_conf(slurm_conf_t *slurm_ctl_conf_ptr);
/*
* slurm_print_ctl_conf - output the contents of slurm control configuration
* message as loaded using slurm_load_ctl_conf
* IN out - file to write to
* IN slurm_ctl_conf_ptr - slurm control configuration pointer
*/
extern void slurm_print_ctl_conf(FILE *out, slurm_conf_t *slurm_ctl_conf_ptr);
/*
* slurm_write_ctl_conf - write the contents of slurm control configuration
* message as loaded using slurm_load_ctl_conf to a file
* IN out - file to write to
* IN slurm_ctl_conf_ptr - slurm control configuration pointer
* IN node_info_ptr - pointer to node table of information
* IN part_info_ptr - pointer to partition information
*/
extern void slurm_write_ctl_conf(slurm_conf_t *slurm_ctl_conf_ptr,
node_info_msg_t *node_info_ptr,
partition_info_msg_t *part_info_ptr);
/*
* slurm_ctl_conf_2_key_pairs - put the slurm_conf_t variables into
* a list_t of opaque data type config_key_pair_t
* IN slurm_ctl_conf_ptr - slurm control configuration pointer
* RET list of opaque data type config_key_pair_t
*/
extern void *slurm_ctl_conf_2_key_pairs(slurm_conf_t *slurm_ctl_conf_ptr);
/*
* slurm_print_key_pairs - output the contents of key_pairs
* which is a list of opaque data type config_key_pair_t
* IN out - file to write to
* IN key_pairs - list containing key pairs to be printed
* IN title - title of key pair list
*/
extern void slurm_print_key_pairs(FILE *out, void *key_pairs, char *title);
/*
* slurm_load_slurmd_status - issue RPC to get the status of slurmd
* daemon on this machine
* IN slurmd_status_ptr - place to store slurmd status information
* RET 0 or -1 on error
* NOTE: free the response using slurm_free_slurmd_status()
*/
extern int slurm_load_slurmd_status(slurmd_status_t **slurmd_status_ptr);
/*
* slurm_free_slurmd_status - free slurmd state information
* IN msg - pointer to slurmd state information
* NOTE: buffer is loaded by slurm_load_slurmd_status
*/
extern void slurm_free_slurmd_status(slurmd_status_t* slurmd_status_ptr);
/*
* slurm_print_slurmd_status - output the contents of slurmd status
* message as loaded using slurm_load_slurmd_status
* IN out - file to write to
* IN slurmd_status_ptr - slurmd status pointer
*/
void slurm_print_slurmd_status(FILE *out, slurmd_status_t *slurmd_status_ptr);
/*
* slurm_init_update_step_msg - initialize step update message with default
* values before calling slurm_update_step()
* OUT step_msg - step update messasge descriptor
*/
extern void slurm_init_update_step_msg(step_update_request_msg_t *step_msg);
/* Get scheduling statistics */
extern int slurm_get_statistics(stats_info_response_msg_t **buf,
stats_info_request_msg_t *req);
/* Reset scheduling statistics */
extern int slurm_reset_statistics(stats_info_request_msg_t *req);
/*****************************************************************************\
* SLURM JOB CONTROL CONFIGURATION READ/PRINT/UPDATE FUNCTIONS
\*****************************************************************************/
/*
* slurm_free_job_info_msg - free the job information response message
* IN msg - pointer to job information response message
* NOTE: buffer is loaded by slurm_load_jobs()
*/
extern void slurm_free_job_info_msg(job_info_msg_t *job_buffer_ptr);
/* Free jobs states response message */
extern void slurm_free_job_state_response_msg(job_state_response_msg_t *msg);
/*
* slurm_free_priority_factors_response_msg - free the job priority factor
* information response message
* IN msg - pointer to job priority information response message
* NOTE: buffer is loaded by slurm_load_job_prio()
*/
extern void slurm_free_priority_factors_response_msg(
priority_factors_response_msg_t *factors_resp);
/*
* slurm_get_end_time - get the expected end time for a given slurm job
* IN jobid - slurm job id
* end_time_ptr - location in which to store scheduled end time for job
* RET 0 or -1 on error
*/
extern int slurm_get_end_time(uint32_t jobid, time_t *end_time_ptr);
/* Given a job record pointer, return its stderr path */
extern void slurm_get_job_stderr(char *buf, int buf_size, job_info_t *job_ptr);
/* Given a job record pointer, return its stdin path */
extern void slurm_get_job_stdin(char *buf, int buf_size, job_info_t *job_ptr);
/* Given a job record pointer, return its stdout path */
extern void slurm_get_job_stdout(char *buf, int buf_size, job_info_t *job_ptr);
/*
* slurm_get_rem_time - get the expected time remaining for a given job
* IN jobid - slurm job id
* RET remaining time in seconds or -1 on error
*/
extern long slurm_get_rem_time(uint32_t jobid);
/*
* slurm_job_node_ready - report if nodes are ready for job to execute now
* IN job_id - slurm job id
* RET: READY_* values defined above
*/
extern int slurm_job_node_ready(uint32_t job_id);
/*
* slurm_load_job - issue RPC to get job information for one job ID
* IN job_info_msg_pptr - place to store a job configuration pointer
* IN job_id - ID of job we want information about
* IN show_flags - job filtering options
* RET 0 or -1 on error
* NOTE: free the response using slurm_free_job_info_msg
*/
extern int slurm_load_job(job_info_msg_t **resp,
uint32_t job_id,
uint16_t show_flags);
/*
* slurm_load_job_prio - issue RPC to get job priority information for jobs
* OUT factors_resp - job priority factors
* IN show_flags - job filtering option: 0 or SHOW_LOCAL
* RET 0 or -1 on error
* NOTE: free the response using slurm_free_priority_factors_response_msg()
*/
extern int slurm_load_job_prio(priority_factors_response_msg_t **factors_resp,
uint16_t show_flags);
/*
* slurm_load_job_user - issue RPC to get slurm information about all jobs
* to be run as the specified user
* IN/OUT job_info_msg_pptr - place to store a job configuration pointer
* IN user_id - ID of user we want information for
* IN show_flags - job filtering options
* RET 0 or -1 on error
* NOTE: free the response using slurm_free_job_info_msg
*/
extern int slurm_load_job_user(job_info_msg_t **job_info_msg_pptr,
uint32_t user_id,
uint16_t show_flags);
/*
* slurm_load_jobs - issue RPC to get slurm all job configuration
* information if changed since update_time
* IN update_time - time of current configuration data
* IN/OUT job_info_msg_pptr - place to store a job configuration pointer
* IN show_flags - job filtering options
* RET 0 or -1 on error
* NOTE: free the response using slurm_free_job_info_msg
*/
extern int slurm_load_jobs(time_t update_time,
job_info_msg_t **job_info_msg_pptr,
uint16_t show_flags);
/*
* slurm_load_job_state - issue RPC to get state of requested jobs
* IN job_id_count - number of jobs in job_ids pointer.
* job_id_count may be 0 to indicate all jobs are requested.
* IN job_ids - ptr to array of jobs with job_ids to query
* IN/OUT jsr_pptr - place to store a query response pointer
* RET SLURM_SUCCESS or error
* NOTE: free the response using slurm_free_job_state_response_msg
*/
extern int slurm_load_job_state(int job_id_count,
slurm_selected_step_t *job_ids,
job_state_response_msg_t **jsr_pptr);
/*
* slurm_notify_job - send message to the job's stdout,
* usable only by user root
* IN job_id - slurm job_id or 0 for all jobs
* IN message - arbitrary message
* RET 0 or -1 on error
*/
extern int slurm_notify_job(uint32_t job_id, char *message);
/*
* slurm_pid2jobid - issue RPC to get the slurm job_id given a process_id
* on this machine
* IN job_pid - process_id of interest on this machine
* OUT job_id_ptr - place to store a slurm job_id
* RET 0 or -1 on error
*/
extern int slurm_pid2jobid(pid_t job_pid, uint32_t *job_id_ptr);
/*
* slurm_update_job - issue RPC to a job's configuration per request,
* only usable by user root or (for some parameters) the job's owner
* IN job_msg - description of job updates
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_update_job(job_desc_msg_t *job_msg);
/*
* slurm_update_job2 - issue RPC to a job's configuration per request,
* only usable by user root or (for some parameters) the job's owner
* IN job_msg - description of job updates
* OUT resp - per task response to the request,
* free using slurm_free_job_array_resp()
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_update_job2(job_desc_msg_t *job_msg,
job_array_resp_msg_t **resp);
/*****************************************************************************\
* SLURM JOB STEP CONFIGURATION READ/PRINT/UPDATE FUNCTIONS
\*****************************************************************************/
/*
* slurm_get_job_steps - issue RPC to get specific slurm job step
* configuration information if changed since update_time.
* a job_id value of NO_VAL implies all jobs, a step_id value of
* NO_VAL implies all steps
* IN update_time - time of current configuration data
* IN job_id - get information for specific job id, NO_VAL for all jobs
* IN step_id - get information for specific job step id, NO_VAL for all
* job steps
* IN step_response_pptr - place to store a step response pointer
* IN show_flags - job step filtering options
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
* NOTE: free the response using slurm_free_job_step_info_response_msg
*/
extern int slurm_get_job_steps(time_t update_time,
uint32_t job_id,
uint32_t step_id,
job_step_info_response_msg_t **step_response_pptr,
uint16_t show_flags);
/*
* Issue RPC to find all steps matching container id and uid (unless uid=NO_VAL)
* IN show_flags - job step filtering options
* IN/OUT steps - List (of slurm_step_id_t*) to populate.
* Must free step ids with slurm_free_step_id().
* RET SLURM_SUCCESS or error
*/
extern int slurm_find_step_ids_by_container_id(uint16_t show_flags, uid_t uid,
const char *container_id,
list_t *steps);
/*
* slurm_free_job_step_info_response_msg - free the job step
* information response message
* IN msg - pointer to job step information response message
* NOTE: buffer is loaded by slurm_get_job_steps.
*/
extern void slurm_free_job_step_info_response_msg(job_step_info_response_msg_t *msg);
/*
* slurm_print_job_step_info_msg - output information about all Slurm
* job steps based upon message as loaded using slurm_get_job_steps
* IN out - file to write to
* IN job_step_info_msg_ptr - job step information message pointer
* IN one_liner - print as a single line if true
*/
extern void slurm_print_job_step_info_msg(FILE *out,
job_step_info_response_msg_t *job_step_info_msg_ptr,
int one_liner);
/*
* slurm_print_job_step_info - output information about a specific Slurm
* job step based upon message as loaded using slurm_get_job_steps
* IN out - file to write to
* IN job_ptr - an individual job step information record pointer
* IN one_liner - print as a single line if true
*/
extern void slurm_print_job_step_info(FILE *out,
job_step_info_t *step_ptr,
int one_liner);
/*
* slurm_job_step_layout_get - get the slurm_step_layout_t structure for
* a particular job step
*
* IN step_id
* RET pointer to a slurm_step_layout_t (free with
* slurm_free_step_layout) on success, and NULL on error.
*/
extern slurm_step_layout_t *slurm_job_step_layout_get(slurm_step_id_t *step_id);
/*
* slurm_sprint_job_step_info - output information about a specific Slurm
* job step based upon message as loaded using slurm_get_job_steps
* IN job_ptr - an individual job step information record pointer
* IN one_liner - print as a single line if true
* RET out - char * containing formatted output (must be freed after call)
* NULL is returned on failure.
*/
extern char *slurm_sprint_job_step_info(job_step_info_t *step_ptr,
int one_liner);
/*
* slurm_job_step_stat - status a current step
*
* IN step_id
* IN node_list, optional, if NULL then all nodes in step are returned.
* OUT resp
* RET SLURM_SUCCESS on success SLURM_ERROR else
*/
extern int slurm_job_step_stat(slurm_step_id_t *step_id,
char *node_list,
uint16_t use_protocol_ver,
job_step_stat_response_msg_t **resp);
/*
* slurm_job_step_get_pids - get the complete list of pids for a given
* job step
*
* IN step_id
* OUT resp
* RET SLURM_SUCCESS on success SLURM_ERROR else
*/
extern int slurm_job_step_get_pids(slurm_step_id_t *step_id,
char *node_list,
job_step_pids_response_msg_t **resp);
extern void slurm_job_step_layout_free(slurm_step_layout_t *layout);
extern void slurm_job_step_pids_free(job_step_pids_t *object);
extern void slurm_job_step_pids_response_msg_free(void *object);
extern void slurm_job_step_stat_free(job_step_stat_t *object);
extern void slurm_job_step_stat_response_msg_free(void *object);
/* Update the time limit of a job step,
* IN step_msg - step update messasge descriptor
* RET 0 or -1 on error */
extern int slurm_update_step(step_update_request_msg_t *step_msg);
extern void slurm_destroy_selected_step(void *object);
/*****************************************************************************\
* SLURM NODE CONFIGURATION READ/PRINT/UPDATE FUNCTIONS
\*****************************************************************************/
/*
* slurm_load_node - issue RPC to get slurm all node configuration information
* if changed since update_time
* IN update_time - time of current configuration data
* OUT resp - place to store a node configuration pointer
* IN show_flags - node filtering options (e.g. SHOW_FEDERATION)
* RET 0 or a slurm error code
* NOTE: free the response using slurm_free_node_info_msg
*/
extern int slurm_load_node(time_t update_time, node_info_msg_t **resp,
uint16_t show_flags);
/*
* slurm_load_node2 - equivalent to slurm_load_node() with addition
* of cluster record for communications in a federation
*/
extern int slurm_load_node2(time_t update_time, node_info_msg_t **resp,
uint16_t show_flags,
slurmdb_cluster_rec_t *cluster);
/*
* slurm_load_node_single - issue RPC to get slurm configuration information
* for a specific node
* OUT resp - place to store a node configuration pointer
* IN node_name - name of the node for which information is requested
* IN show_flags - node filtering options
* RET 0 or a slurm error code
* NOTE: free the response using slurm_free_node_info_msg
*/
extern int slurm_load_node_single(node_info_msg_t **resp, char *node_name,
uint16_t show_flags);
/*
* slurm_load_node_single2 - equivalent to slurm_load_node_single() with
* addition of cluster record for communications in a federation
*/
extern int slurm_load_node_single2(node_info_msg_t **resp, char *node_name,
uint16_t show_flags,
slurmdb_cluster_rec_t *cluster);
/* Given data structures containing information about nodes and partitions,
* populate the node's "partitions" field */
void
slurm_populate_node_partitions(node_info_msg_t *node_buffer_ptr,
partition_info_msg_t *part_buffer_ptr);
/*
* slurm_get_node_energy - issue RPC to get the energy data of all
* configured sensors on the target machine
* IN host - name of node to query, NULL if localhost
* IN delta - Use cache if data is newer than this in seconds
* OUT sensor_cnt - number of sensors
* OUT energy - array of acct_gather_energy_t structures on success or
* NULL other wise
* RET 0 on success or a slurm error code
* NOTE: free the response using xfree
*/
extern int slurm_get_node_energy(char *host,
uint16_t context_id,
uint16_t delta,
uint16_t *sensors_cnt,
acct_gather_energy_t **energy);
extern int slurm_get_node_alias_addrs(char *node_list,
slurm_node_alias_addrs_t **alias_addrs);
/*
* slurm_free_node_info_msg - free the node information response message
* IN msg - pointer to node information response message
* NOTE: buffer is loaded by slurm_load_node.
*/
extern void slurm_free_node_info_msg(node_info_msg_t *node_buffer_ptr);
/*
* slurm_print_node_info_msg - output information about all Slurm nodes
* based upon message as loaded using slurm_load_node
* IN out - file to write to
* IN node_info_msg_ptr - node information message pointer
* IN one_liner - print as a single line if true
*/
extern void slurm_print_node_info_msg(FILE *out,
node_info_msg_t *node_info_msg_ptr,
int one_liner);
/*
* slurm_print_node_table - output information about a specific Slurm nodes
* based upon message as loaded using slurm_load_node
* IN out - file to write to
* IN node_ptr - an individual node information record pointer
* IN one_liner - print as a single line if true
*/
extern void slurm_print_node_table(FILE *out,
node_info_t *node_ptr,
int one_liner);
/*
* slurm_sprint_node_table - output information about a specific Slurm nodes
* based upon message as loaded using slurm_load_node
* IN node_ptr - an individual node information record pointer
* IN one_liner - print as a single line if true
* RET out - char * containing formatted output (must be freed after call)
* NULL is returned on failure.
*/
extern char *slurm_sprint_node_table(node_info_t *node_ptr,
int one_liner);
/*
* slurm_init_update_node_msg - initialize node update message
* OUT update_node_msg - user defined node descriptor
*/
void slurm_init_update_node_msg(update_node_msg_t *update_node_msg);
/*
* slurm_create_node - issue RPC to create node(s), only usable by user root
* IN node_msg - node definition(s)
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_create_node(update_node_msg_t *node_msg);
/*
* slurm_update_node - issue RPC to a node's configuration per request,
* only usable by user root
* IN node_msg - description of node updates
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_update_node(update_node_msg_t *node_msg);
/*
* slurm_delete_node - issue RPC to delete a node, only usable by user root
* IN node_msg - use to pass nodelist of names to delete
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
int slurm_delete_node(update_node_msg_t *node_msg);
/*****************************************************************************\
* SLURM FRONT_END CONFIGURATION READ/PRINT/UPDATE FUNCTIONS
\*****************************************************************************/
/*
* slurm_load_front_end - issue RPC to get slurm all front_end configuration
* information if changed since update_time
* IN update_time - time of current configuration data
* IN front_end_info_msg_pptr - place to store a front_end configuration pointer
* RET 0 or a slurm error code
* NOTE: free the response using slurm_free_front_end_info_msg
*/
extern int slurm_load_front_end(time_t update_time,
front_end_info_msg_t **resp);
/*
* slurm_free_front_end_info_msg - free the front_end information response
* message
* IN msg - pointer to front_end information response message
* NOTE: buffer is loaded by slurm_load_front_end.
*/
extern void slurm_free_front_end_info_msg(front_end_info_msg_t *front_end_buffer_ptr);
/*
* slurm_print_front_end_info_msg - output information about all Slurm
* front_ends based upon message as loaded using slurm_load_front_end
* IN out - file to write to
* IN front_end_info_msg_ptr - front_end information message pointer
* IN one_liner - print as a single line if true
*/
extern void slurm_print_front_end_info_msg(FILE *out,
front_end_info_msg_t *front_end_info_msg_ptr,
int one_liner);
/*
* slurm_print_front_end_table - output information about a specific Slurm
* front_ends based upon message as loaded using slurm_load_front_end
* IN out - file to write to
* IN front_end_ptr - an individual front_end information record pointer
* IN one_liner - print as a single line if true
*/
extern void slurm_print_front_end_table(FILE *out,
front_end_info_t *front_end_ptr,
int one_liner);
/*
* slurm_sprint_front_end_table - output information about a specific Slurm
* front_end based upon message as loaded using slurm_load_front_end
* IN front_end_ptr - an individual front_end information record pointer
* IN one_liner - print as a single line if true
* RET out - char * containing formatted output (must be freed after call)
* NULL is returned on failure.
*/
extern char *slurm_sprint_front_end_table(front_end_info_t *front_end_ptr,
int one_liner);
/*
* slurm_init_update_front_end_msg - initialize front_end node update message
* OUT update_front_end_msg - user defined node descriptor
*/
void slurm_init_update_front_end_msg(update_front_end_msg_t *update_front_end_msg);
/*
* slurm_update_front_end - issue RPC to a front_end node's configuration per
* request, only usable by user root
* IN front_end_msg - description of front_end node updates
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_update_front_end(update_front_end_msg_t *front_end_msg);
/*****************************************************************************\
* SLURM SWITCH TOPOLOGY CONFIGURATION READ/PRINT FUNCTIONS
\*****************************************************************************/
/*
* slurm_load_topo - issue RPC to get slurm all switch topology configuration
* information
* IN node_info_msg_pptr - place to store a node configuration pointer
* RET 0 or a slurm error code
* NOTE: free the response using slurm_free_topo_info_msg
*/
extern int slurm_load_topo(topo_info_response_msg_t **topo_info_msg_pptr);
/*
* slurm_free_topo_info_msg - free the switch topology configuration
* information response message
* IN msg - pointer to switch topology configuration response message
* NOTE: buffer is loaded by slurm_load_topo.
*/
extern void slurm_free_topo_info_msg(topo_info_response_msg_t *msg);
/*
* slurm_print_topo_info_msg - output information about all switch topology
* configuration information based upon message as loaded using
* slurm_load_topo
* IN out - file to write to
* IN topo_info_msg_ptr - switch topology information message pointer
* IN node_list - NULL to print all topology information
* IN one_liner - print as a single line if not zero
*/
extern void slurm_print_topo_info_msg(FILE *out,
topo_info_response_msg_t *topo_info_msg_ptr,
char *node_list, int one_liner);
/*****************************************************************************\
* SLURM SELECT READ/PRINT/UPDATE FUNCTIONS
\*****************************************************************************/
/*
* slurm_get_select_nodeinfo - get data from a select node credential
* IN nodeinfo - updated select node credential
* IN data_type - type of data to enter into node credential
* IN state - state of node query
* IN/OUT data - the data to enter into node credential
* RET 0 or -1 on error
*/
extern int slurm_get_select_nodeinfo(dynamic_plugin_data_t *nodeinfo,
enum select_nodedata_type data_type,
enum node_states state,
void *data);
/*****************************************************************************\
* SLURM PARTITION CONFIGURATION READ/PRINT/UPDATE FUNCTIONS
\*****************************************************************************/
/*
* slurm_init_part_desc_msg - initialize partition descriptor with
* default values
* IN/OUT update_part_msg - user defined partition descriptor
*/
extern void slurm_init_part_desc_msg(update_part_msg_t *update_part_msg);
/*
* slurm_load_partitions - issue RPC to get slurm all partition configuration
* information if changed since update_time
* IN update_time - time of current configuration data
* IN partition_info_msg_pptr - place to store a partition configuration
* pointer
* IN show_flags - partitions filtering options (e.g. SHOW_FEDERATION)
* RET 0 or a slurm error code
* NOTE: free the response using slurm_free_partition_info_msg
*/
extern int slurm_load_partitions(time_t update_time,
partition_info_msg_t **part_buffer_ptr,
uint16_t show_flags);
/*
* slurm_load_partitions2 - equivalent to slurm_load_partitions() with addition
* of cluster record for communications in a federation
*/
extern int slurm_load_partitions2(time_t update_time,
partition_info_msg_t **resp,
uint16_t show_flags,
slurmdb_cluster_rec_t *cluster);
/*
* slurm_free_partition_info_msg - free the partition information
* response message
* IN msg - pointer to partition information response message
* NOTE: buffer is loaded by slurm_load_partitions
*/
extern void slurm_free_partition_info_msg(partition_info_msg_t *part_info_ptr);
/*
* slurm_print_partition_info_msg - output information about all Slurm
* partitions based upon message as loaded using slurm_load_partitions
* IN out - file to write to
* IN part_info_ptr - partitions information message pointer
* IN one_liner - print as a single line if true
*/
extern void slurm_print_partition_info_msg(FILE *out, partition_info_msg_t *part_info_ptr, int one_liner);
/*
* slurm_print_partition_info - output information about a specific Slurm
* partition based upon message as loaded using slurm_load_partitions
* IN out - file to write to
* IN part_ptr - an individual partition information record pointer
* IN one_liner - print as a single line if true
*/
extern void slurm_print_partition_info(FILE *out,
partition_info_t *part_ptr,
int one_liner);
/*
* slurm_sprint_partition_info - output information about a specific Slurm
* partition based upon message as loaded using slurm_load_partitions
* IN part_ptr - an individual partition information record pointer
* IN one_liner - print as a single line if true
* RET out - char * with formatted output (must be freed after call)
* NULL is returned on failure.
*/
extern char *slurm_sprint_partition_info(partition_info_t *part_ptr,
int one_liner);
/*
* slurm_create_partition - create a new partition, only usable by user root
* IN part_msg - description of partition configuration
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_create_partition(update_part_msg_t *part_msg);
/*
* slurm_update_partition - issue RPC to update a partition's configuration
* per request, only usable by user root
* IN part_msg - description of partition updates
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_update_partition(update_part_msg_t *part_msg);
/*
* slurm_delete_partition - issue RPC to delete a partition, only usable
* by user root
* IN part_msg - description of partition to delete
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_delete_partition(delete_part_msg_t *part_msg);
/*****************************************************************************\
* SLURM RESERVATION CONFIGURATION READ/PRINT/UPDATE FUNCTIONS
\*****************************************************************************/
/*
* slurm_init_resv_desc_msg - initialize reservation descriptor with
* default values
* OUT job_desc_msg - user defined partition descriptor
*/
extern void slurm_init_resv_desc_msg(resv_desc_msg_t *update_resv_msg);
/*
* slurm_create_reservation - create a new reservation, only usable by user root
* IN resv_msg - description of reservation
* RET name of reservation on success (caller must free the memory),
* otherwise return NULL and set errno to indicate the error
*/
extern char *slurm_create_reservation(resv_desc_msg_t *resv_msg);
/*
* slurm_update_reservation - modify an existing reservation, only usable by
* user root
* IN resv_msg - description of reservation
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_update_reservation(resv_desc_msg_t *resv_msg);
/*
* slurm_delete_reservation - issue RPC to delete a reservation, only usable
* by user root
* IN resv_msg - description of reservation to delete
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_delete_reservation(reservation_name_msg_t *resv_msg);
/*
* slurm_load_reservations - issue RPC to get all slurm reservation
* configuration information if changed since update_time
* IN update_time - time of current configuration data
* IN reserve_info_msg_pptr - place to store a reservation configuration
* pointer
* RET 0 or a slurm error code
* NOTE: free the response using slurm_free_reservation_info_msg
*/
extern int slurm_load_reservations(time_t update_time,
reserve_info_msg_t **resp);
/*
* slurm_print_reservation_info_msg - output information about all Slurm
* reservations based upon message as loaded using slurm_load_reservation
* IN out - file to write to
* IN resv_info_ptr - reservation information message pointer
* IN one_liner - print as a single line if true
*/
void slurm_print_reservation_info_msg(FILE* out,
reserve_info_msg_t *resv_info_ptr,
int one_liner);
/*
* slurm_print_reservation_info - output information about a specific Slurm
* reservation based upon message as loaded using slurm_load_reservation
* IN out - file to write to
* IN resv_ptr - an individual reservation information record pointer
* IN one_liner - print as a single line if true
*/
void slurm_print_reservation_info(FILE* out,
reserve_info_t *resv_ptr,
int one_liner);
/*
* slurm_sprint_reservation_info - output information about a specific Slurm
* reservation based upon message as loaded using slurm_load_reservations
* IN resv_ptr - an individual reservation information record pointer
* IN one_liner - print as a single line if true
* RET out - char * containing formatted output (must be freed after call)
* NULL is returned on failure.
*/
char *slurm_sprint_reservation_info(reserve_info_t *resv_ptr, int one_liner);
/*
* slurm_free_reservation_info_msg - free the reservation information
* response message
* IN msg - pointer to reservation information response message
* NOTE: buffer is loaded by slurm_load_reservation
*/
extern void slurm_free_reservation_info_msg(reserve_info_msg_t *resv_info_ptr);
/*****************************************************************************\
* SLURM PING/RECONFIGURE/SHUTDOWN STRUCTURES
\*****************************************************************************/
typedef struct {
char *hostname; /* symlink - do not xfree() */
bool pinged; /* true on successful ping */
long latency; /* time to ping or timeout on !pinged */
/*
* controller offset which defines default mode:
* 0: primary
* 1: backup
* 2+: backup#
*/
int offset;
} controller_ping_t;
typedef struct {
char *hostname; /* symlink - do not xfree() */
bool pinged; /* true on successful ping */
long latency; /* time to ping or timeout on !pinged */
int offset; /* offset which defines default mode:
* 0: primary
* 1: backup
*/
} slurmdbd_ping_t;
/*****************************************************************************\
* SLURM PING/RECONFIGURE/SHUTDOWN FUNCTIONS
\*****************************************************************************/
/*
* slurm_ping - issue RPC to have Slurm controller (slurmctld)
* IN dest - controller to contact (0=primary, 1=backup, 2=backup2, etc.)
* RET 0 or a slurm error code
*/
extern int slurm_ping(int dest);
/*
* RET array of each ping result (NULL terminated).
* Caller must xfree() the result.
*/
extern controller_ping_t *ping_all_controllers();
/*
* slurm_reconfigure - issue RPC to have Slurm controller (slurmctld)
* reload its configuration file
* RET 0 or a slurm error code
*/
extern int slurm_reconfigure(void);
/*
* slurm_shutdown - issue RPC to have Slurm controller (slurmctld)
* cease operations, both the primary and all backup controllers
* are shutdown.
* IN options - 0: all slurm daemons are shutdown
* 1: slurmctld generates a core file
* 2: only the slurmctld is shutdown (no core file)
* RET 0 or a slurm error code
*/
extern int slurm_shutdown(uint16_t options);
/*
* slurm_takeover - issue RPC to have a Slurm backup controller take over the
* primary controller. REQUEST_CONTROL is sent by the backup
* to the primary controller to take control
* backup_inx IN - Index of BackupController to assume controller (typically 1)
* RET 0 or a slurm error code
*/
extern int slurm_takeover(int backup_inx);
/*
* slurm_set_debugflags - issue RPC to set slurm controller debug flags
* IN debug_flags_plus - debug flags to be added
* IN debug_flags_minus - debug flags to be removed
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_set_debugflags(uint64_t debug_flags_plus,
uint64_t debug_flags_minus);
/*
* slurm_set_slurmd_debug_flags - issue RPC to set slurmd debug flags
* IN debug_flags_plus - debug flags to be added
* IN debug_flags_minus - debug flags to be removed
* IN debug_flags_set - new debug flags value
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR
*/
extern int slurm_set_slurmd_debug_flags(char *node_list,
uint64_t debug_flags_plus,
uint64_t debug_flags_minus);
/*
* slurm_set_slurmd_debug_level - issue RPC to set slurmd debug level
* IN debug_level - requested debug level
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR
*/
extern int slurm_set_slurmd_debug_level(char *node_list, uint32_t debug_level);
/*
* slurm_set_debug_level - issue RPC to set slurm controller debug level
* IN debug_level - requested debug level
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_set_debug_level(uint32_t debug_level);
/*
* slurm_set_schedlog_level - issue RPC to set slurm scheduler log level
* IN schedlog_level - requested scheduler log level
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_set_schedlog_level(uint32_t schedlog_level);
/*
* slurm_set_fs_dampeningfactor - issue RPC to set slurm fs dampening factor
* IN factor - requested fs dampening factor
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_set_fs_dampeningfactor(uint16_t factor);
/*
* slurm_update_suspend_exc_nodes - issue RPC to set SuspendExcNodes
* IN nodes - string to set
* IN mode - Whether to set, append or remove nodes from the setting
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_update_suspend_exc_nodes(char *nodes, update_mode_t mode);
/*
* slurm_update_suspend_exc_parts - issue RPC to set SuspendExcParts
* IN parts - string to set
* IN mode - Whether to set, append or remove partitions from the setting
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_update_suspend_exc_parts(char *parts, update_mode_t mode);
/*
* slurm_update_suspend_exc_states - issue RPC to set SuspendExcStates
* IN states - string to set
* IN mode - Whether to set, append or remove states from the setting
* RET SLURM_SUCCESS on success, otherwise return SLURM_ERROR with errno set
*/
extern int slurm_update_suspend_exc_states(char *states, update_mode_t mode);
/*****************************************************************************\
* SLURM JOB SUSPEND FUNCTIONS
\*****************************************************************************/
/*
* slurm_suspend - suspend execution of a job.
* IN job_id - job on which to perform operation
* RET 0 or a slurm error code
*/
extern int slurm_suspend(uint32_t job_id);
/*
* slurm_suspend2 - suspend execution of a job.
* IN job_id in string form - job on which to perform operation, may be job
* array specification (e.g. "123_1-20,44");
* OUT resp - per task response to the request,
* free using slurm_free_job_array_resp()
* RET 0 or a slurm error code
*/
extern int slurm_suspend2(char *job_id, job_array_resp_msg_t **resp);
/*
* slurm_resume - resume execution of a previously suspended job.
* IN job_id - job on which to perform operation
* RET 0 or a slurm error code
*/
extern int slurm_resume(uint32_t job_id);
/*
* slurm_resume2 - resume execution of a previously suspended job.
* IN job_id in string form - job on which to perform operation, may be job
* array specification (e.g. "123_1-20,44");
* OUT resp - per task response to the request,
* free using slurm_free_job_array_resp()
* RET 0 or a slurm error code
*/
extern int slurm_resume2(char *job_id, job_array_resp_msg_t **resp);
/* Free job array oriented response with individual return codes by task ID */
extern void slurm_free_job_array_resp(job_array_resp_msg_t *resp);
/*
* slurm_requeue - re-queue a batch job, if already running
* then terminate it first
* IN job_id - job on which to perform operation
* IN flags - JOB_SPECIAL_EXIT - job should be placed special exit state and
* held.
* JOB_REQUEUE_HOLD - job should be placed JOB_PENDING state and
* held.
* JOB_RECONFIG_FAIL - Node configuration for job failed
* JOB_RUNNING - Operate only on jobs in a state of
* CONFIGURING, RUNNING, STOPPED or SUSPENDED.
* RET 0 or a slurm error code
*/
extern int slurm_requeue(uint32_t job_id, uint32_t flags);
/*
* slurm_requeue2 - re-queue a batch job, if already running
* then terminate it first
* IN job_id in string form - job on which to perform operation, may be job
* array specification (e.g. "123_1-20,44");
* IN flags - JOB_SPECIAL_EXIT - job should be placed special exit state and
* held.
* JOB_REQUEUE_HOLD - job should be placed JOB_PENDING state and
* held.
* JOB_RECONFIG_FAIL - Node configuration for job failed
* JOB_RUNNING - Operate only on jobs in a state of
* CONFIGURING, RUNNING, STOPPED or SUSPENDED.
* OUT resp - per task response to the request,
* free using slurm_free_job_array_resp()
* RET 0 or a slurm error code
*/
extern int slurm_requeue2(char *job_id, uint32_t flags,
job_array_resp_msg_t **resp);
/*****************************************************************************\
* SLURM TRIGGER FUNCTIONS
\*****************************************************************************/
/*
* slurm_set_trigger - Set an event trigger
* RET 0 or a slurm error code
*/
extern int slurm_set_trigger(trigger_info_t *trigger_set);
/*
* slurm_clear_trigger - Clear (remove) an existing event trigger
* RET 0 or a slurm error code
*/
extern int slurm_clear_trigger(trigger_info_t *trigger_clear);
/*
* slurm_get_triggers - Get all event trigger information
* Use slurm_free_trigger_msg() to free the memory allocated by this function
* RET 0 or a slurm error code
*/
extern int slurm_get_triggers(trigger_info_msg_t **trigger_get);
/*
* slurm_pull_trigger - Pull an event trigger
* RET 0 or a slurm error code
*/
extern int slurm_pull_trigger(trigger_info_t *trigger_pull);
/*
* slurm_free_trigger_msg - Free data structure returned by
* slurm_get_triggers()
*/
extern void slurm_free_trigger_msg(trigger_info_msg_t *trigger_free);
/*
* slurm_init_trigger_msg - initialize trigger clear/update message
* OUT trigger_info_msg - user defined trigger descriptor
*/
void slurm_init_trigger_msg(trigger_info_t *trigger_info_msg);
/*****************************************************************************\
* SLURM BURST BUFFER FUNCTIONS
\*****************************************************************************/
#define BB_FLAG_DISABLE_PERSISTENT 0x0001 /* Disable regular user to create
* and destroy persistent burst
* buffers */
#define BB_FLAG_ENABLE_PERSISTENT 0x0002 /* Allow regular user to create
* and destroy persistent burst
* buffers */
#define BB_FLAG_EMULATE_CRAY 0x0004 /* Using dw_wlm_cli emulator */
#define BB_FLAG_PRIVATE_DATA 0x0008 /* Buffers only visible to owner */
#define BB_FLAG_TEARDOWN_FAILURE 0x0010 /* Teardown after failed staged in/out */
#define BB_SIZE_IN_NODES 0x8000000000000000
/*
* Burst buffer states: Keep in sync with bb_state_string() and bb_state_num()
* in slurm_protocol_defs.c.
*/
#define BB_STATE_PENDING 0x0000 /* Placeholder: no action started */
#define BB_STATE_ALLOCATING 0x0001 /* Cray: bbs_setup started */
#define BB_STATE_ALLOCATED 0x0002 /* Cray: bbs_setup started */
#define BB_STATE_DELETING 0x0005 /* Cray: bbs_setup started */
#define BB_STATE_DELETED 0x0006 /* Cray: bbs_setup started */
#define BB_STATE_STAGING_IN 0x0011 /* Cray: bbs_data_in started */
#define BB_STATE_STAGED_IN 0x0012 /* Cray: bbs_data_in complete */
#define BB_STATE_PRE_RUN 0x0018 /* Cray: bbs_pre_run started */
#define BB_STATE_ALLOC_REVOKE 0x001a /* Cray: allocation revoked */
#define BB_STATE_RUNNING 0x0021 /* Job is running */
#define BB_STATE_SUSPEND 0x0022 /* Job is suspended (future) */
#define BB_STATE_POST_RUN 0x0029 /* Cray: bbs_post_run started */
#define BB_STATE_STAGING_OUT 0x0031 /* Cray: bbs_data_out started */
#define BB_STATE_STAGED_OUT 0x0032 /* Cray: bbs_data_out complete */
#define BB_STATE_TEARDOWN 0x0041 /* Cray: bbs_teardown started */
#define BB_STATE_TEARDOWN_FAIL 0x0043 /* Cray: bbs_teardown failed, retrying */
#define BB_STATE_COMPLETE 0x0045 /* Cray: bbs_teardown complete */
/* Information about alternate pools or other burst buffer resources */
typedef struct {
uint64_t granularity; /* Granularity of resource allocation size */
char *name; /* Resource (pool) name */
uint64_t total_space; /* Total size of available resources, unused
* by burst_buffer_resv_t */
uint64_t used_space; /* Allocated space, in bytes */
uint64_t unfree_space; /* used plus drained space, units are bytes */
} burst_buffer_pool_t;
typedef struct {
char *account; /* Associated account (for limits) */
uint32_t array_job_id;
uint32_t array_task_id;
time_t create_time; /* Time of creation */
uint32_t job_id;
char *name; /* Name of persistent burst buffer */
char *partition; /* Associated partition (for limits) */
char *pool; /* Resource (pool) name */
char *qos; /* Associated QOS (for limits) */
uint64_t size; /* In bytes by default */
uint16_t state; /* See BB_STATE_* */
uint32_t user_id;
} burst_buffer_resv_t;
typedef struct {
uint32_t user_id;
uint64_t used;
} burst_buffer_use_t;
typedef struct {
char *allow_users;
char *default_pool; /* Name of default pool to use */
char *create_buffer;
char *deny_users;
char *destroy_buffer;
uint32_t flags; /* See BB_FLAG_* above */
char *get_sys_state;
char *get_sys_status;
uint64_t granularity; /* Granularity of resource allocation */
uint32_t pool_cnt; /* Count of records in pool_ptr */
burst_buffer_pool_t *pool_ptr;
char *name; /* Plugin name */
uint32_t poll_interval;
uint32_t other_timeout; /* Seconds or zero */
uint32_t stage_in_timeout; /* Seconds or zero */
uint32_t stage_out_timeout; /* Seconds or zero */
char *start_stage_in;
char *start_stage_out;
char *stop_stage_in;
char *stop_stage_out;
uint64_t total_space; /* In bytes */
uint64_t unfree_space; /* Allocated or drained, in bytes */
uint64_t used_space; /* Allocated, in bytes */
uint32_t validate_timeout; /* Seconds or zero */
uint32_t buffer_count;
burst_buffer_resv_t *burst_buffer_resv_ptr;
uint32_t use_count;
burst_buffer_use_t *burst_buffer_use_ptr;
} burst_buffer_info_t;
typedef struct {
burst_buffer_info_t *burst_buffer_array;
uint32_t record_count; /* Elements in burst_buffer_array */
} burst_buffer_info_msg_t;
/*
* slurm_burst_buffer_state_string - translate burst buffer state number to
* it string equivalent
*/
extern char *slurm_burst_buffer_state_string(uint16_t state);
/*
* slurm_load_burst_buffer_stat - issue RPC to get burst buffer status
* IN argc - count of status request options
* IN argv - status request options
* OUT status_resp - status response, memory must be released using xfree()
* RET 0 or a slurm error code
*/
extern int slurm_load_burst_buffer_stat(int argc, char **argv,
char **status_resp);
/*
* slurm_load_burst_buffer_info - issue RPC to get slurm all burst buffer plugin
* information
* OUT burst_buffer_info_msg_pptr - place to store a burst buffer configuration
* pointer
* RET 0 or a slurm error code
* NOTE: free the response using slurm_free_burst_buffer_info_msg
*/
extern int slurm_load_burst_buffer_info(burst_buffer_info_msg_t **burst_buffer_info_msg_pptr);
/*
* slurm_free_burst_buffer_info_msg - free buffer returned by
* slurm_load_burst_buffer
* IN burst_buffer_info_msg_ptr - pointer to burst_buffer_info_msg_t
* RET 0 or a slurm error code
*/
extern void slurm_free_burst_buffer_info_msg(burst_buffer_info_msg_t *burst_buffer_info_msg);
/*
* slurm_print_burst_buffer_info_msg - output information about burst buffers
* based upon message as loaded using slurm_load_burst_buffer
* IN out - file to write to
* IN info_ptr - burst_buffer information message pointer
* IN one_liner - print as a single line if true
* IN verbose - higher values to log additional details
*/
extern void slurm_print_burst_buffer_info_msg(FILE *out,
burst_buffer_info_msg_t *info_ptr,
int one_liner,
int verbosity);
/*
* slurm_print_burst_buffer_record - output information about a specific Slurm
* burst_buffer record based upon message as loaded using
* slurm_load_burst_buffer_info()
* IN out - file to write to
* IN burst_buffer_ptr - an individual burst buffer record pointer
* IN one_liner - print as a single line if not zero
* IN verbose - higher values to log additional details
* RET out - char * containing formatted output (must be freed after call)
* NULL is returned on failure.
*/
extern void slurm_print_burst_buffer_record(FILE *out,
burst_buffer_info_t *burst_buffer_ptr,
int one_liner,
int verbose);
/*
* slurm_network_callerid - issue RPC to get the job id of a job from a remote
* slurmd based upon network socket information.
*
* IN req - Information about network connection in question
* OUT job_id - ID of the job or NO_VAL
* OUT node_name - name of the remote slurmd
* IN node_name_size - size of the node_name buffer
* RET SLURM_SUCCESS or SLURM_ERROR on error
*/
extern int slurm_network_callerid(network_callerid_msg_t req,
uint32_t *job_id,
char *node_name,
int node_name_size);
/*
* Move the specified job ID to the top of the queue for a given user ID,
* partition, account, and QOS.
* IN job_id_str - a job id
* RET 0 or -1 on error */
extern int slurm_top_job(char *job_id_str);
/*
* Fetch an auth token for a given username.
* IN username - NULL, or a specific username if run as SlurmUser/root.
* IN lifespan - lifespan the token should be valid for.
*/
extern char *slurm_fetch_token(char *username, int lifespan);
/*****************************************************************************\
* SLURM FEDERATION FUNCTIONS
\*****************************************************************************/
/*
* slurm_load_federation - issue RPC to get federation status from controller
* IN/OUT fed_pptr - place to store returned federation information.
* slurmdb_federation_rec_t treated as a void pointer to since
* slurm.h doesn't have ties to slurmdb.h.
* NOTE: Use slurm_destroy_federation_rec() to release the returned memory
* RET 0 or -1 on error
*/
extern int slurm_load_federation(void **fed_pptr);
/*
* slurm_print_federation - prints slurmdb_federation_rec_t (passed as void*
* since slurm.h doesn't know about slurmdb.h).
*/
extern void slurm_print_federation(void *fed);
/*
* slurm_destroy_federation_rec - Release memory allocated by
* slurm_load_federation()
*/
extern void slurm_destroy_federation_rec(void *fed);
/*****************************************************************************\
* SLURM CRONTAB FUNCTIONS
\*****************************************************************************/
extern int slurm_request_crontab(uid_t uid, char **crontab,
char **disabled_lines);
typedef struct {
char *err_msg;
char *failed_lines;
uint32_t *jobids;
uint32_t jobids_count;
char *job_submit_user_msg;
uint32_t return_code;
} crontab_update_response_msg_t;
extern crontab_update_response_msg_t *slurm_update_crontab(uid_t uid, gid_t gid,
char *crontab,
list_t *jobs);
extern int slurm_remove_crontab(uid_t uid, gid_t gid);
#ifdef __cplusplus
}
#endif
#endif
|