File: daemon_unix.go

package info (click to toggle)
docker.io 18.09.5+dfsg1-1
  • links: PTS, VCS
  • area: main
  • in suites: experimental
  • size: 65,324 kB
  • sloc: sh: 9,757; makefile: 833; ansic: 239; python: 162; asm: 10
file content (1523 lines) | stat: -rw-r--r-- 57,161 bytes parent folder | download | duplicates (5)
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
// +build linux freebsd

package daemon // import "github.com/docker/docker/daemon"

import (
	"bufio"
	"context"
	"fmt"
	"io/ioutil"
	"net"
	"os"
	"path/filepath"
	"runtime"
	"runtime/debug"
	"strconv"
	"strings"
	"time"

	containerd_cgroups "github.com/containerd/cgroups"
	"github.com/docker/docker/api/types"
	"github.com/docker/docker/api/types/blkiodev"
	pblkiodev "github.com/docker/docker/api/types/blkiodev"
	containertypes "github.com/docker/docker/api/types/container"
	"github.com/docker/docker/container"
	"github.com/docker/docker/daemon/config"
	"github.com/docker/docker/daemon/initlayer"
	"github.com/docker/docker/opts"
	"github.com/docker/docker/pkg/containerfs"
	"github.com/docker/docker/pkg/idtools"
	"github.com/docker/docker/pkg/ioutils"
	"github.com/docker/docker/pkg/mount"
	"github.com/docker/docker/pkg/parsers"
	"github.com/docker/docker/pkg/parsers/kernel"
	"github.com/docker/docker/pkg/sysinfo"
	"github.com/docker/docker/runconfig"
	volumemounts "github.com/docker/docker/volume/mounts"
	"github.com/docker/libnetwork"
	nwconfig "github.com/docker/libnetwork/config"
	"github.com/docker/libnetwork/drivers/bridge"
	"github.com/docker/libnetwork/netlabel"
	"github.com/docker/libnetwork/netutils"
	"github.com/docker/libnetwork/options"
	lntypes "github.com/docker/libnetwork/types"
	"github.com/opencontainers/runc/libcontainer/cgroups"
	rsystem "github.com/opencontainers/runc/libcontainer/system"
	"github.com/opencontainers/runtime-spec/specs-go"
	"github.com/opencontainers/selinux/go-selinux/label"
	"github.com/pkg/errors"
	"github.com/sirupsen/logrus"
	"github.com/vishvananda/netlink"
	"golang.org/x/sys/unix"
)

const (
	// DefaultShimBinary is the default shim to be used by containerd if none
	// is specified
	DefaultShimBinary = "docker-containerd-shim"

	// DefaultRuntimeBinary is the default runtime to be used by
	// containerd if none is specified
	DefaultRuntimeBinary = "runc"

	// See https://git.kernel.org/cgit/linux/kernel/git/tip/tip.git/tree/kernel/sched/sched.h?id=8cd9234c64c584432f6992fe944ca9e46ca8ea76#n269
	linuxMinCPUShares = 2
	linuxMaxCPUShares = 262144
	platformSupported = true
	// It's not kernel limit, we want this 4M limit to supply a reasonable functional container
	linuxMinMemory = 4194304
	// constants for remapped root settings
	defaultIDSpecifier = "default"
	defaultRemappedID  = "dockremap"

	// constant for cgroup drivers
	cgroupFsDriver      = "cgroupfs"
	cgroupSystemdDriver = "systemd"

	// DefaultRuntimeName is the default runtime to be used by
	// containerd if none is specified
	DefaultRuntimeName = "runc"
)

type containerGetter interface {
	GetContainer(string) (*container.Container, error)
}

func getMemoryResources(config containertypes.Resources) *specs.LinuxMemory {
	memory := specs.LinuxMemory{}

	if config.Memory > 0 {
		memory.Limit = &config.Memory
	}

	if config.MemoryReservation > 0 {
		memory.Reservation = &config.MemoryReservation
	}

	if config.MemorySwap > 0 {
		memory.Swap = &config.MemorySwap
	}

	if config.MemorySwappiness != nil {
		swappiness := uint64(*config.MemorySwappiness)
		memory.Swappiness = &swappiness
	}

	if config.OomKillDisable != nil {
		memory.DisableOOMKiller = config.OomKillDisable
	}

	if config.KernelMemory != 0 {
		memory.Kernel = &config.KernelMemory
	}

	return &memory
}

func getCPUResources(config containertypes.Resources) (*specs.LinuxCPU, error) {
	cpu := specs.LinuxCPU{}

	if config.CPUShares < 0 {
		return nil, fmt.Errorf("shares: invalid argument")
	}
	if config.CPUShares >= 0 {
		shares := uint64(config.CPUShares)
		cpu.Shares = &shares
	}

	if config.CpusetCpus != "" {
		cpu.Cpus = config.CpusetCpus
	}

	if config.CpusetMems != "" {
		cpu.Mems = config.CpusetMems
	}

	if config.NanoCPUs > 0 {
		// https://www.kernel.org/doc/Documentation/scheduler/sched-bwc.txt
		period := uint64(100 * time.Millisecond / time.Microsecond)
		quota := config.NanoCPUs * int64(period) / 1e9
		cpu.Period = &period
		cpu.Quota = &quota
	}

	if config.CPUPeriod != 0 {
		period := uint64(config.CPUPeriod)
		cpu.Period = &period
	}

	if config.CPUQuota != 0 {
		q := config.CPUQuota
		cpu.Quota = &q
	}

	if config.CPURealtimePeriod != 0 {
		period := uint64(config.CPURealtimePeriod)
		cpu.RealtimePeriod = &period
	}

	if config.CPURealtimeRuntime != 0 {
		c := config.CPURealtimeRuntime
		cpu.RealtimeRuntime = &c
	}

	return &cpu, nil
}

func getBlkioWeightDevices(config containertypes.Resources) ([]specs.LinuxWeightDevice, error) {
	var stat unix.Stat_t
	var blkioWeightDevices []specs.LinuxWeightDevice

	for _, weightDevice := range config.BlkioWeightDevice {
		if err := unix.Stat(weightDevice.Path, &stat); err != nil {
			return nil, err
		}
		weight := weightDevice.Weight
		d := specs.LinuxWeightDevice{Weight: &weight}
		d.Major = int64(stat.Rdev / 256)
		d.Minor = int64(stat.Rdev % 256)
		blkioWeightDevices = append(blkioWeightDevices, d)
	}

	return blkioWeightDevices, nil
}

func (daemon *Daemon) parseSecurityOpt(container *container.Container, hostConfig *containertypes.HostConfig) error {
	container.NoNewPrivileges = daemon.configStore.NoNewPrivileges
	return parseSecurityOpt(container, hostConfig)
}

func parseSecurityOpt(container *container.Container, config *containertypes.HostConfig) error {
	var (
		labelOpts []string
		err       error
	)

	for _, opt := range config.SecurityOpt {
		if opt == "no-new-privileges" {
			container.NoNewPrivileges = true
			continue
		}
		if opt == "disable" {
			labelOpts = append(labelOpts, "disable")
			continue
		}

		var con []string
		if strings.Contains(opt, "=") {
			con = strings.SplitN(opt, "=", 2)
		} else if strings.Contains(opt, ":") {
			con = strings.SplitN(opt, ":", 2)
			logrus.Warn("Security options with `:` as a separator are deprecated and will be completely unsupported in 17.04, use `=` instead.")
		}
		if len(con) != 2 {
			return fmt.Errorf("invalid --security-opt 1: %q", opt)
		}

		switch con[0] {
		case "label":
			labelOpts = append(labelOpts, con[1])
		case "apparmor":
			container.AppArmorProfile = con[1]
		case "seccomp":
			container.SeccompProfile = con[1]
		case "no-new-privileges":
			noNewPrivileges, err := strconv.ParseBool(con[1])
			if err != nil {
				return fmt.Errorf("invalid --security-opt 2: %q", opt)
			}
			container.NoNewPrivileges = noNewPrivileges
		default:
			return fmt.Errorf("invalid --security-opt 2: %q", opt)
		}
	}

	container.ProcessLabel, container.MountLabel, err = label.InitLabels(labelOpts)
	return err
}

func getBlkioThrottleDevices(devs []*blkiodev.ThrottleDevice) ([]specs.LinuxThrottleDevice, error) {
	var throttleDevices []specs.LinuxThrottleDevice
	var stat unix.Stat_t

	for _, d := range devs {
		if err := unix.Stat(d.Path, &stat); err != nil {
			return nil, err
		}
		d := specs.LinuxThrottleDevice{Rate: d.Rate}
		d.Major = int64(stat.Rdev / 256)
		d.Minor = int64(stat.Rdev % 256)
		throttleDevices = append(throttleDevices, d)
	}

	return throttleDevices, nil
}

func checkKernel() error {
	// Check for unsupported kernel versions
	// FIXME: it would be cleaner to not test for specific versions, but rather
	// test for specific functionalities.
	// Unfortunately we can't test for the feature "does not cause a kernel panic"
	// without actually causing a kernel panic, so we need this workaround until
	// the circumstances of pre-3.10 crashes are clearer.
	// For details see https://github.com/docker/docker/issues/407
	// Docker 1.11 and above doesn't actually run on kernels older than 3.4,
	// due to containerd-shim usage of PR_SET_CHILD_SUBREAPER (introduced in 3.4).
	if !kernel.CheckKernelVersion(3, 10, 0) {
		v, _ := kernel.GetKernelVersion()
		if os.Getenv("DOCKER_NOWARN_KERNEL_VERSION") == "" {
			logrus.Fatalf("Your Linux kernel version %s is not supported for running docker. Please upgrade your kernel to 3.10.0 or newer.", v.String())
		}
	}
	return nil
}

// adaptContainerSettings is called during container creation to modify any
// settings necessary in the HostConfig structure.
func (daemon *Daemon) adaptContainerSettings(hostConfig *containertypes.HostConfig, adjustCPUShares bool) error {
	if adjustCPUShares && hostConfig.CPUShares > 0 {
		// Handle unsupported CPUShares
		if hostConfig.CPUShares < linuxMinCPUShares {
			logrus.Warnf("Changing requested CPUShares of %d to minimum allowed of %d", hostConfig.CPUShares, linuxMinCPUShares)
			hostConfig.CPUShares = linuxMinCPUShares
		} else if hostConfig.CPUShares > linuxMaxCPUShares {
			logrus.Warnf("Changing requested CPUShares of %d to maximum allowed of %d", hostConfig.CPUShares, linuxMaxCPUShares)
			hostConfig.CPUShares = linuxMaxCPUShares
		}
	}
	if hostConfig.Memory > 0 && hostConfig.MemorySwap == 0 {
		// By default, MemorySwap is set to twice the size of Memory.
		hostConfig.MemorySwap = hostConfig.Memory * 2
	}
	if hostConfig.ShmSize == 0 {
		hostConfig.ShmSize = config.DefaultShmSize
		if daemon.configStore != nil {
			hostConfig.ShmSize = int64(daemon.configStore.ShmSize)
		}
	}
	// Set default IPC mode, if unset for container
	if hostConfig.IpcMode.IsEmpty() {
		m := config.DefaultIpcMode
		if daemon.configStore != nil {
			m = daemon.configStore.IpcMode
		}
		hostConfig.IpcMode = containertypes.IpcMode(m)
	}

	adaptSharedNamespaceContainer(daemon, hostConfig)

	var err error
	opts, err := daemon.generateSecurityOpt(hostConfig)
	if err != nil {
		return err
	}
	hostConfig.SecurityOpt = append(hostConfig.SecurityOpt, opts...)
	if hostConfig.OomKillDisable == nil {
		defaultOomKillDisable := false
		hostConfig.OomKillDisable = &defaultOomKillDisable
	}

	return nil
}

// adaptSharedNamespaceContainer replaces container name with its ID in hostConfig.
// To be more precisely, it modifies `container:name` to `container:ID` of PidMode, IpcMode
// and NetworkMode.
//
// When a container shares its namespace with another container, use ID can keep the namespace
// sharing connection between the two containers even the another container is renamed.
func adaptSharedNamespaceContainer(daemon containerGetter, hostConfig *containertypes.HostConfig) {
	containerPrefix := "container:"
	if hostConfig.PidMode.IsContainer() {
		pidContainer := hostConfig.PidMode.Container()
		// if there is any error returned here, we just ignore it and leave it to be
		// handled in the following logic
		if c, err := daemon.GetContainer(pidContainer); err == nil {
			hostConfig.PidMode = containertypes.PidMode(containerPrefix + c.ID)
		}
	}
	if hostConfig.IpcMode.IsContainer() {
		ipcContainer := hostConfig.IpcMode.Container()
		if c, err := daemon.GetContainer(ipcContainer); err == nil {
			hostConfig.IpcMode = containertypes.IpcMode(containerPrefix + c.ID)
		}
	}
	if hostConfig.NetworkMode.IsContainer() {
		netContainer := hostConfig.NetworkMode.ConnectedContainer()
		if c, err := daemon.GetContainer(netContainer); err == nil {
			hostConfig.NetworkMode = containertypes.NetworkMode(containerPrefix + c.ID)
		}
	}
}

func verifyContainerResources(resources *containertypes.Resources, sysInfo *sysinfo.SysInfo, update bool) ([]string, error) {
	warnings := []string{}
	fixMemorySwappiness(resources)

	// memory subsystem checks and adjustments
	if resources.Memory != 0 && resources.Memory < linuxMinMemory {
		return warnings, fmt.Errorf("Minimum memory limit allowed is 4MB")
	}
	if resources.Memory > 0 && !sysInfo.MemoryLimit {
		warnings = append(warnings, "Your kernel does not support memory limit capabilities or the cgroup is not mounted. Limitation discarded.")
		logrus.Warn("Your kernel does not support memory limit capabilities or the cgroup is not mounted. Limitation discarded.")
		resources.Memory = 0
		resources.MemorySwap = -1
	}
	if resources.Memory > 0 && resources.MemorySwap != -1 && !sysInfo.SwapLimit {
		warnings = append(warnings, "Your kernel does not support swap limit capabilities or the cgroup is not mounted. Memory limited without swap.")
		logrus.Warn("Your kernel does not support swap limit capabilities,or the cgroup is not mounted. Memory limited without swap.")
		resources.MemorySwap = -1
	}
	if resources.Memory > 0 && resources.MemorySwap > 0 && resources.MemorySwap < resources.Memory {
		return warnings, fmt.Errorf("Minimum memoryswap limit should be larger than memory limit, see usage")
	}
	if resources.Memory == 0 && resources.MemorySwap > 0 && !update {
		return warnings, fmt.Errorf("You should always set the Memory limit when using Memoryswap limit, see usage")
	}
	if resources.MemorySwappiness != nil && !sysInfo.MemorySwappiness {
		warnings = append(warnings, "Your kernel does not support memory swappiness capabilities or the cgroup is not mounted. Memory swappiness discarded.")
		logrus.Warn("Your kernel does not support memory swappiness capabilities, or the cgroup is not mounted. Memory swappiness discarded.")
		resources.MemorySwappiness = nil
	}
	if resources.MemorySwappiness != nil {
		swappiness := *resources.MemorySwappiness
		if swappiness < 0 || swappiness > 100 {
			return warnings, fmt.Errorf("Invalid value: %v, valid memory swappiness range is 0-100", swappiness)
		}
	}
	if resources.MemoryReservation > 0 && !sysInfo.MemoryReservation {
		warnings = append(warnings, "Your kernel does not support memory soft limit capabilities or the cgroup is not mounted. Limitation discarded.")
		logrus.Warn("Your kernel does not support memory soft limit capabilities or the cgroup is not mounted. Limitation discarded.")
		resources.MemoryReservation = 0
	}
	if resources.MemoryReservation > 0 && resources.MemoryReservation < linuxMinMemory {
		return warnings, fmt.Errorf("Minimum memory reservation allowed is 4MB")
	}
	if resources.Memory > 0 && resources.MemoryReservation > 0 && resources.Memory < resources.MemoryReservation {
		return warnings, fmt.Errorf("Minimum memory limit can not be less than memory reservation limit, see usage")
	}
	if resources.KernelMemory > 0 && !sysInfo.KernelMemory {
		warnings = append(warnings, "Your kernel does not support kernel memory limit capabilities or the cgroup is not mounted. Limitation discarded.")
		logrus.Warn("Your kernel does not support kernel memory limit capabilities or the cgroup is not mounted. Limitation discarded.")
		resources.KernelMemory = 0
	}
	if resources.KernelMemory > 0 && resources.KernelMemory < linuxMinMemory {
		return warnings, fmt.Errorf("Minimum kernel memory limit allowed is 4MB")
	}
	if resources.KernelMemory > 0 && !kernel.CheckKernelVersion(4, 0, 0) {
		warnings = append(warnings, "You specified a kernel memory limit on a kernel older than 4.0. Kernel memory limits are experimental on older kernels, it won't work as expected and can cause your system to be unstable.")
		logrus.Warn("You specified a kernel memory limit on a kernel older than 4.0. Kernel memory limits are experimental on older kernels, it won't work as expected and can cause your system to be unstable.")
	}
	if resources.OomKillDisable != nil && !sysInfo.OomKillDisable {
		// only produce warnings if the setting wasn't to *disable* the OOM Kill; no point
		// warning the caller if they already wanted the feature to be off
		if *resources.OomKillDisable {
			warnings = append(warnings, "Your kernel does not support OomKillDisable. OomKillDisable discarded.")
			logrus.Warn("Your kernel does not support OomKillDisable. OomKillDisable discarded.")
		}
		resources.OomKillDisable = nil
	}

	if resources.PidsLimit != 0 && !sysInfo.PidsLimit {
		warnings = append(warnings, "Your kernel does not support pids limit capabilities or the cgroup is not mounted. PIDs limit discarded.")
		logrus.Warn("Your kernel does not support pids limit capabilities or the cgroup is not mounted. PIDs limit discarded.")
		resources.PidsLimit = 0
	}

	// cpu subsystem checks and adjustments
	if resources.NanoCPUs > 0 && resources.CPUPeriod > 0 {
		return warnings, fmt.Errorf("Conflicting options: Nano CPUs and CPU Period cannot both be set")
	}
	if resources.NanoCPUs > 0 && resources.CPUQuota > 0 {
		return warnings, fmt.Errorf("Conflicting options: Nano CPUs and CPU Quota cannot both be set")
	}
	if resources.NanoCPUs > 0 && (!sysInfo.CPUCfsPeriod || !sysInfo.CPUCfsQuota) {
		return warnings, fmt.Errorf("NanoCPUs can not be set, as your kernel does not support CPU cfs period/quota or the cgroup is not mounted")
	}
	// The highest precision we could get on Linux is 0.001, by setting
	//   cpu.cfs_period_us=1000ms
	//   cpu.cfs_quota=1ms
	// See the following link for details:
	// https://www.kernel.org/doc/Documentation/scheduler/sched-bwc.txt
	// Here we don't set the lower limit and it is up to the underlying platform (e.g., Linux) to return an error.
	// The error message is 0.01 so that this is consistent with Windows
	if resources.NanoCPUs < 0 || resources.NanoCPUs > int64(sysinfo.NumCPU())*1e9 {
		return warnings, fmt.Errorf("Range of CPUs is from 0.01 to %d.00, as there are only %d CPUs available", sysinfo.NumCPU(), sysinfo.NumCPU())
	}

	if resources.CPUShares > 0 && !sysInfo.CPUShares {
		warnings = append(warnings, "Your kernel does not support CPU shares or the cgroup is not mounted. Shares discarded.")
		logrus.Warn("Your kernel does not support CPU shares or the cgroup is not mounted. Shares discarded.")
		resources.CPUShares = 0
	}
	if resources.CPUPeriod > 0 && !sysInfo.CPUCfsPeriod {
		warnings = append(warnings, "Your kernel does not support CPU cfs period or the cgroup is not mounted. Period discarded.")
		logrus.Warn("Your kernel does not support CPU cfs period or the cgroup is not mounted. Period discarded.")
		resources.CPUPeriod = 0
	}
	if resources.CPUPeriod != 0 && (resources.CPUPeriod < 1000 || resources.CPUPeriod > 1000000) {
		return warnings, fmt.Errorf("CPU cfs period can not be less than 1ms (i.e. 1000) or larger than 1s (i.e. 1000000)")
	}
	if resources.CPUQuota > 0 && !sysInfo.CPUCfsQuota {
		warnings = append(warnings, "Your kernel does not support CPU cfs quota or the cgroup is not mounted. Quota discarded.")
		logrus.Warn("Your kernel does not support CPU cfs quota or the cgroup is not mounted. Quota discarded.")
		resources.CPUQuota = 0
	}
	if resources.CPUQuota > 0 && resources.CPUQuota < 1000 {
		return warnings, fmt.Errorf("CPU cfs quota can not be less than 1ms (i.e. 1000)")
	}
	if resources.CPUPercent > 0 {
		warnings = append(warnings, fmt.Sprintf("%s does not support CPU percent. Percent discarded.", runtime.GOOS))
		logrus.Warnf("%s does not support CPU percent. Percent discarded.", runtime.GOOS)
		resources.CPUPercent = 0
	}

	// cpuset subsystem checks and adjustments
	if (resources.CpusetCpus != "" || resources.CpusetMems != "") && !sysInfo.Cpuset {
		warnings = append(warnings, "Your kernel does not support cpuset or the cgroup is not mounted. Cpuset discarded.")
		logrus.Warn("Your kernel does not support cpuset or the cgroup is not mounted. Cpuset discarded.")
		resources.CpusetCpus = ""
		resources.CpusetMems = ""
	}
	cpusAvailable, err := sysInfo.IsCpusetCpusAvailable(resources.CpusetCpus)
	if err != nil {
		return warnings, errors.Wrapf(err, "Invalid value %s for cpuset cpus", resources.CpusetCpus)
	}
	if !cpusAvailable {
		return warnings, fmt.Errorf("Requested CPUs are not available - requested %s, available: %s", resources.CpusetCpus, sysInfo.Cpus)
	}
	memsAvailable, err := sysInfo.IsCpusetMemsAvailable(resources.CpusetMems)
	if err != nil {
		return warnings, errors.Wrapf(err, "Invalid value %s for cpuset mems", resources.CpusetMems)
	}
	if !memsAvailable {
		return warnings, fmt.Errorf("Requested memory nodes are not available - requested %s, available: %s", resources.CpusetMems, sysInfo.Mems)
	}

	// blkio subsystem checks and adjustments
	if resources.BlkioWeight > 0 && !sysInfo.BlkioWeight {
		warnings = append(warnings, "Your kernel does not support Block I/O weight or the cgroup is not mounted. Weight discarded.")
		logrus.Warn("Your kernel does not support Block I/O weight or the cgroup is not mounted. Weight discarded.")
		resources.BlkioWeight = 0
	}
	if resources.BlkioWeight > 0 && (resources.BlkioWeight < 10 || resources.BlkioWeight > 1000) {
		return warnings, fmt.Errorf("Range of blkio weight is from 10 to 1000")
	}
	if resources.IOMaximumBandwidth != 0 || resources.IOMaximumIOps != 0 {
		return warnings, fmt.Errorf("Invalid QoS settings: %s does not support Maximum IO Bandwidth or Maximum IO IOps", runtime.GOOS)
	}
	if len(resources.BlkioWeightDevice) > 0 && !sysInfo.BlkioWeightDevice {
		warnings = append(warnings, "Your kernel does not support Block I/O weight_device or the cgroup is not mounted. Weight-device discarded.")
		logrus.Warn("Your kernel does not support Block I/O weight_device or the cgroup is not mounted. Weight-device discarded.")
		resources.BlkioWeightDevice = []*pblkiodev.WeightDevice{}
	}
	if len(resources.BlkioDeviceReadBps) > 0 && !sysInfo.BlkioReadBpsDevice {
		warnings = append(warnings, "Your kernel does not support BPS Block I/O read limit or the cgroup is not mounted. Block I/O BPS read limit discarded.")
		logrus.Warn("Your kernel does not support BPS Block I/O read limit or the cgroup is not mounted. Block I/O BPS read limit discarded")
		resources.BlkioDeviceReadBps = []*pblkiodev.ThrottleDevice{}
	}
	if len(resources.BlkioDeviceWriteBps) > 0 && !sysInfo.BlkioWriteBpsDevice {
		warnings = append(warnings, "Your kernel does not support BPS Block I/O write limit or the cgroup is not mounted. Block I/O BPS write limit discarded.")
		logrus.Warn("Your kernel does not support BPS Block I/O write limit or the cgroup is not mounted. Block I/O BPS write limit discarded.")
		resources.BlkioDeviceWriteBps = []*pblkiodev.ThrottleDevice{}

	}
	if len(resources.BlkioDeviceReadIOps) > 0 && !sysInfo.BlkioReadIOpsDevice {
		warnings = append(warnings, "Your kernel does not support IOPS Block read limit or the cgroup is not mounted. Block I/O IOPS read limit discarded.")
		logrus.Warn("Your kernel does not support IOPS Block I/O read limit in IO or the cgroup is not mounted. Block I/O IOPS read limit discarded.")
		resources.BlkioDeviceReadIOps = []*pblkiodev.ThrottleDevice{}
	}
	if len(resources.BlkioDeviceWriteIOps) > 0 && !sysInfo.BlkioWriteIOpsDevice {
		warnings = append(warnings, "Your kernel does not support IOPS Block write limit or the cgroup is not mounted. Block I/O IOPS write limit discarded.")
		logrus.Warn("Your kernel does not support IOPS Block I/O write limit or the cgroup is not mounted. Block I/O IOPS write limit discarded.")
		resources.BlkioDeviceWriteIOps = []*pblkiodev.ThrottleDevice{}
	}

	return warnings, nil
}

func (daemon *Daemon) getCgroupDriver() string {
	cgroupDriver := cgroupFsDriver

	if UsingSystemd(daemon.configStore) {
		cgroupDriver = cgroupSystemdDriver
	}
	return cgroupDriver
}

// getCD gets the raw value of the native.cgroupdriver option, if set.
func getCD(config *config.Config) string {
	for _, option := range config.ExecOptions {
		key, val, err := parsers.ParseKeyValueOpt(option)
		if err != nil || !strings.EqualFold(key, "native.cgroupdriver") {
			continue
		}
		return val
	}
	return ""
}

// VerifyCgroupDriver validates native.cgroupdriver
func VerifyCgroupDriver(config *config.Config) error {
	cd := getCD(config)
	if cd == "" || cd == cgroupFsDriver || cd == cgroupSystemdDriver {
		return nil
	}
	return fmt.Errorf("native.cgroupdriver option %s not supported", cd)
}

// UsingSystemd returns true if cli option includes native.cgroupdriver=systemd
func UsingSystemd(config *config.Config) bool {
	return getCD(config) == cgroupSystemdDriver
}

// verifyPlatformContainerSettings performs platform-specific validation of the
// hostconfig and config structures.
func verifyPlatformContainerSettings(daemon *Daemon, hostConfig *containertypes.HostConfig, config *containertypes.Config, update bool) ([]string, error) {
	var warnings []string
	sysInfo := sysinfo.New(true)

	w, err := verifyContainerResources(&hostConfig.Resources, sysInfo, update)

	// no matter err is nil or not, w could have data in itself.
	warnings = append(warnings, w...)

	if err != nil {
		return warnings, err
	}

	if hostConfig.ShmSize < 0 {
		return warnings, fmt.Errorf("SHM size can not be less than 0")
	}

	if hostConfig.OomScoreAdj < -1000 || hostConfig.OomScoreAdj > 1000 {
		return warnings, fmt.Errorf("Invalid value %d, range for oom score adj is [-1000, 1000]", hostConfig.OomScoreAdj)
	}

	// ip-forwarding does not affect container with '--net=host' (or '--net=none')
	if sysInfo.IPv4ForwardingDisabled && !(hostConfig.NetworkMode.IsHost() || hostConfig.NetworkMode.IsNone()) {
		warnings = append(warnings, "IPv4 forwarding is disabled. Networking will not work.")
		logrus.Warn("IPv4 forwarding is disabled. Networking will not work")
	}
	// check for various conflicting options with user namespaces
	if daemon.configStore.RemappedRoot != "" && hostConfig.UsernsMode.IsPrivate() {
		if hostConfig.Privileged {
			return warnings, fmt.Errorf("privileged mode is incompatible with user namespaces.  You must run the container in the host namespace when running privileged mode")
		}
		if hostConfig.NetworkMode.IsHost() && !hostConfig.UsernsMode.IsHost() {
			return warnings, fmt.Errorf("cannot share the host's network namespace when user namespaces are enabled")
		}
		if hostConfig.PidMode.IsHost() && !hostConfig.UsernsMode.IsHost() {
			return warnings, fmt.Errorf("cannot share the host PID namespace when user namespaces are enabled")
		}
	}
	if hostConfig.CgroupParent != "" && UsingSystemd(daemon.configStore) {
		// CgroupParent for systemd cgroup should be named as "xxx.slice"
		if len(hostConfig.CgroupParent) <= 6 || !strings.HasSuffix(hostConfig.CgroupParent, ".slice") {
			return warnings, fmt.Errorf("cgroup-parent for systemd cgroup should be a valid slice named as \"xxx.slice\"")
		}
	}
	if hostConfig.Runtime == "" {
		hostConfig.Runtime = daemon.configStore.GetDefaultRuntimeName()
	}

	if rt := daemon.configStore.GetRuntime(hostConfig.Runtime); rt == nil {
		return warnings, fmt.Errorf("Unknown runtime specified %s", hostConfig.Runtime)
	}

	parser := volumemounts.NewParser(runtime.GOOS)
	for dest := range hostConfig.Tmpfs {
		if err := parser.ValidateTmpfsMountDestination(dest); err != nil {
			return warnings, err
		}
	}

	return warnings, nil
}

func (daemon *Daemon) loadRuntimes() error {
	return daemon.initRuntimes(daemon.configStore.Runtimes)
}

func (daemon *Daemon) initRuntimes(runtimes map[string]types.Runtime) (err error) {
	runtimeDir := filepath.Join(daemon.configStore.Root, "runtimes")
	// Remove old temp directory if any
	os.RemoveAll(runtimeDir + "-old")
	tmpDir, err := ioutils.TempDir(daemon.configStore.Root, "gen-runtimes")
	if err != nil {
		return errors.Wrap(err, "failed to get temp dir to generate runtime scripts")
	}
	defer func() {
		if err != nil {
			if err1 := os.RemoveAll(tmpDir); err1 != nil {
				logrus.WithError(err1).WithField("dir", tmpDir).
					Warn("failed to remove tmp dir")
			}
			return
		}

		if err = os.Rename(runtimeDir, runtimeDir+"-old"); err != nil {
			return
		}
		if err = os.Rename(tmpDir, runtimeDir); err != nil {
			err = errors.Wrap(err, "failed to setup runtimes dir, new containers may not start")
			return
		}
		if err = os.RemoveAll(runtimeDir + "-old"); err != nil {
			logrus.WithError(err).WithField("dir", tmpDir).
				Warn("failed to remove old runtimes dir")
		}
	}()

	for name, rt := range runtimes {
		if len(rt.Args) == 0 {
			continue
		}

		script := filepath.Join(tmpDir, name)
		content := fmt.Sprintf("#!/bin/sh\n%s %s $@\n", rt.Path, strings.Join(rt.Args, " "))
		if err := ioutil.WriteFile(script, []byte(content), 0700); err != nil {
			return err
		}
	}
	return nil
}

// verifyDaemonSettings performs validation of daemon config struct
func verifyDaemonSettings(conf *config.Config) error {
	// Check for mutually incompatible config options
	if conf.BridgeConfig.Iface != "" && conf.BridgeConfig.IP != "" {
		return fmt.Errorf("You specified -b & --bip, mutually exclusive options. Please specify only one")
	}
	if !conf.BridgeConfig.EnableIPTables && !conf.BridgeConfig.InterContainerCommunication {
		return fmt.Errorf("You specified --iptables=false with --icc=false. ICC=false uses iptables to function. Please set --icc or --iptables to true")
	}
	if !conf.BridgeConfig.EnableIPTables && conf.BridgeConfig.EnableIPMasq {
		conf.BridgeConfig.EnableIPMasq = false
	}
	if err := VerifyCgroupDriver(conf); err != nil {
		return err
	}
	if conf.CgroupParent != "" && UsingSystemd(conf) {
		if len(conf.CgroupParent) <= 6 || !strings.HasSuffix(conf.CgroupParent, ".slice") {
			return fmt.Errorf("cgroup-parent for systemd cgroup should be a valid slice named as \"xxx.slice\"")
		}
	}

	if conf.DefaultRuntime == "" {
		conf.DefaultRuntime = config.StockRuntimeName
	}
	if conf.Runtimes == nil {
		conf.Runtimes = make(map[string]types.Runtime)
	}
	conf.Runtimes[config.StockRuntimeName] = types.Runtime{Path: DefaultRuntimeName}

	return nil
}

// checkSystem validates platform-specific requirements
func checkSystem() error {
	if os.Geteuid() != 0 {
		return fmt.Errorf("The Docker daemon needs to be run as root")
	}
	return checkKernel()
}

// configureMaxThreads sets the Go runtime max threads threshold
// which is 90% of the kernel setting from /proc/sys/kernel/threads-max
func configureMaxThreads(config *config.Config) error {
	mt, err := ioutil.ReadFile("/proc/sys/kernel/threads-max")
	if err != nil {
		return err
	}
	mtint, err := strconv.Atoi(strings.TrimSpace(string(mt)))
	if err != nil {
		return err
	}
	maxThreads := (mtint / 100) * 90
	debug.SetMaxThreads(maxThreads)
	logrus.Debugf("Golang's threads limit set to %d", maxThreads)
	return nil
}

func overlaySupportsSelinux() (bool, error) {
	f, err := os.Open("/proc/kallsyms")
	if err != nil {
		if os.IsNotExist(err) {
			return false, nil
		}
		return false, err
	}
	defer f.Close()

	var symAddr, symType, symName, text string

	s := bufio.NewScanner(f)
	for s.Scan() {
		if err := s.Err(); err != nil {
			return false, err
		}

		text = s.Text()
		if _, err := fmt.Sscanf(text, "%s %s %s", &symAddr, &symType, &symName); err != nil {
			return false, fmt.Errorf("Scanning '%s' failed: %s", text, err)
		}

		// Check for presence of symbol security_inode_copy_up.
		if symName == "security_inode_copy_up" {
			return true, nil
		}
	}
	return false, nil
}

// configureKernelSecuritySupport configures and validates security support for the kernel
func configureKernelSecuritySupport(config *config.Config, driverName string) error {
	if config.EnableSelinuxSupport {
		if !selinuxEnabled() {
			logrus.Warn("Docker could not enable SELinux on the host system")
			return nil
		}

		if driverName == "overlay" || driverName == "overlay2" {
			// If driver is overlay or overlay2, make sure kernel
			// supports selinux with overlay.
			supported, err := overlaySupportsSelinux()
			if err != nil {
				return err
			}

			if !supported {
				logrus.Warnf("SELinux is not supported with the %v graph driver on this kernel", driverName)
			}
		}
	} else {
		selinuxSetDisabled()
	}
	return nil
}

func (daemon *Daemon) initNetworkController(config *config.Config, activeSandboxes map[string]interface{}) (libnetwork.NetworkController, error) {
	netOptions, err := daemon.networkOptions(config, daemon.PluginStore, activeSandboxes)
	if err != nil {
		return nil, err
	}

	controller, err := libnetwork.New(netOptions...)
	if err != nil {
		return nil, fmt.Errorf("error obtaining controller instance: %v", err)
	}

	if len(activeSandboxes) > 0 {
		logrus.Info("There are old running containers, the network config will not take affect")
		return controller, nil
	}

	// Initialize default network on "null"
	if n, _ := controller.NetworkByName("none"); n == nil {
		if _, err := controller.NewNetwork("null", "none", "", libnetwork.NetworkOptionPersist(true)); err != nil {
			return nil, fmt.Errorf("Error creating default \"null\" network: %v", err)
		}
	}

	// Initialize default network on "host"
	if n, _ := controller.NetworkByName("host"); n == nil {
		if _, err := controller.NewNetwork("host", "host", "", libnetwork.NetworkOptionPersist(true)); err != nil {
			return nil, fmt.Errorf("Error creating default \"host\" network: %v", err)
		}
	}

	// Clear stale bridge network
	if n, err := controller.NetworkByName("bridge"); err == nil {
		if err = n.Delete(); err != nil {
			return nil, fmt.Errorf("could not delete the default bridge network: %v", err)
		}
		if len(config.NetworkConfig.DefaultAddressPools.Value()) > 0 && !daemon.configStore.LiveRestoreEnabled {
			removeDefaultBridgeInterface()
		}
	}

	if !config.DisableBridge {
		// Initialize default driver "bridge"
		if err := initBridgeDriver(controller, config); err != nil {
			return nil, err
		}
	} else {
		removeDefaultBridgeInterface()
	}

	return controller, nil
}

func driverOptions(config *config.Config) []nwconfig.Option {
	bridgeConfig := options.Generic{
		"EnableIPForwarding":  config.BridgeConfig.EnableIPForward,
		"EnableIPTables":      config.BridgeConfig.EnableIPTables,
		"EnableUserlandProxy": config.BridgeConfig.EnableUserlandProxy,
		"UserlandProxyPath":   config.BridgeConfig.UserlandProxyPath}
	bridgeOption := options.Generic{netlabel.GenericData: bridgeConfig}

	dOptions := []nwconfig.Option{}
	dOptions = append(dOptions, nwconfig.OptionDriverConfig("bridge", bridgeOption))
	return dOptions
}

func initBridgeDriver(controller libnetwork.NetworkController, config *config.Config) error {
	bridgeName := bridge.DefaultBridgeName
	if config.BridgeConfig.Iface != "" {
		bridgeName = config.BridgeConfig.Iface
	}
	netOption := map[string]string{
		bridge.BridgeName:         bridgeName,
		bridge.DefaultBridge:      strconv.FormatBool(true),
		netlabel.DriverMTU:        strconv.Itoa(config.Mtu),
		bridge.EnableIPMasquerade: strconv.FormatBool(config.BridgeConfig.EnableIPMasq),
		bridge.EnableICC:          strconv.FormatBool(config.BridgeConfig.InterContainerCommunication),
	}

	// --ip processing
	if config.BridgeConfig.DefaultIP != nil {
		netOption[bridge.DefaultBindingIP] = config.BridgeConfig.DefaultIP.String()
	}

	var (
		ipamV4Conf *libnetwork.IpamConf
		ipamV6Conf *libnetwork.IpamConf
	)

	ipamV4Conf = &libnetwork.IpamConf{AuxAddresses: make(map[string]string)}

	nwList, nw6List, err := netutils.ElectInterfaceAddresses(bridgeName)
	if err != nil {
		return errors.Wrap(err, "list bridge addresses failed")
	}

	nw := nwList[0]
	if len(nwList) > 1 && config.BridgeConfig.FixedCIDR != "" {
		_, fCIDR, err := net.ParseCIDR(config.BridgeConfig.FixedCIDR)
		if err != nil {
			return errors.Wrap(err, "parse CIDR failed")
		}
		// Iterate through in case there are multiple addresses for the bridge
		for _, entry := range nwList {
			if fCIDR.Contains(entry.IP) {
				nw = entry
				break
			}
		}
	}

	ipamV4Conf.PreferredPool = lntypes.GetIPNetCanonical(nw).String()
	hip, _ := lntypes.GetHostPartIP(nw.IP, nw.Mask)
	if hip.IsGlobalUnicast() {
		ipamV4Conf.Gateway = nw.IP.String()
	}

	if config.BridgeConfig.IP != "" {
		ipamV4Conf.PreferredPool = config.BridgeConfig.IP
		ip, _, err := net.ParseCIDR(config.BridgeConfig.IP)
		if err != nil {
			return err
		}
		ipamV4Conf.Gateway = ip.String()
	} else if bridgeName == bridge.DefaultBridgeName && ipamV4Conf.PreferredPool != "" {
		logrus.Infof("Default bridge (%s) is assigned with an IP address %s. Daemon option --bip can be used to set a preferred IP address", bridgeName, ipamV4Conf.PreferredPool)
	}

	if config.BridgeConfig.FixedCIDR != "" {
		_, fCIDR, err := net.ParseCIDR(config.BridgeConfig.FixedCIDR)
		if err != nil {
			return err
		}

		ipamV4Conf.SubPool = fCIDR.String()
	}

	if config.BridgeConfig.DefaultGatewayIPv4 != nil {
		ipamV4Conf.AuxAddresses["DefaultGatewayIPv4"] = config.BridgeConfig.DefaultGatewayIPv4.String()
	}

	var deferIPv6Alloc bool
	if config.BridgeConfig.FixedCIDRv6 != "" {
		_, fCIDRv6, err := net.ParseCIDR(config.BridgeConfig.FixedCIDRv6)
		if err != nil {
			return err
		}

		// In case user has specified the daemon flag --fixed-cidr-v6 and the passed network has
		// at least 48 host bits, we need to guarantee the current behavior where the containers'
		// IPv6 addresses will be constructed based on the containers' interface MAC address.
		// We do so by telling libnetwork to defer the IPv6 address allocation for the endpoints
		// on this network until after the driver has created the endpoint and returned the
		// constructed address. Libnetwork will then reserve this address with the ipam driver.
		ones, _ := fCIDRv6.Mask.Size()
		deferIPv6Alloc = ones <= 80

		if ipamV6Conf == nil {
			ipamV6Conf = &libnetwork.IpamConf{AuxAddresses: make(map[string]string)}
		}
		ipamV6Conf.PreferredPool = fCIDRv6.String()

		// In case the --fixed-cidr-v6 is specified and the current docker0 bridge IPv6
		// address belongs to the same network, we need to inform libnetwork about it, so
		// that it can be reserved with IPAM and it will not be given away to somebody else
		for _, nw6 := range nw6List {
			if fCIDRv6.Contains(nw6.IP) {
				ipamV6Conf.Gateway = nw6.IP.String()
				break
			}
		}
	}

	if config.BridgeConfig.DefaultGatewayIPv6 != nil {
		if ipamV6Conf == nil {
			ipamV6Conf = &libnetwork.IpamConf{AuxAddresses: make(map[string]string)}
		}
		ipamV6Conf.AuxAddresses["DefaultGatewayIPv6"] = config.BridgeConfig.DefaultGatewayIPv6.String()
	}

	v4Conf := []*libnetwork.IpamConf{ipamV4Conf}
	v6Conf := []*libnetwork.IpamConf{}
	if ipamV6Conf != nil {
		v6Conf = append(v6Conf, ipamV6Conf)
	}
	// Initialize default network on "bridge" with the same name
	_, err = controller.NewNetwork("bridge", "bridge", "",
		libnetwork.NetworkOptionEnableIPv6(config.BridgeConfig.EnableIPv6),
		libnetwork.NetworkOptionDriverOpts(netOption),
		libnetwork.NetworkOptionIpam("default", "", v4Conf, v6Conf, nil),
		libnetwork.NetworkOptionDeferIPv6Alloc(deferIPv6Alloc))
	if err != nil {
		return fmt.Errorf("Error creating default \"bridge\" network: %v", err)
	}
	return nil
}

// Remove default bridge interface if present (--bridge=none use case)
func removeDefaultBridgeInterface() {
	if lnk, err := netlink.LinkByName(bridge.DefaultBridgeName); err == nil {
		if err := netlink.LinkDel(lnk); err != nil {
			logrus.Warnf("Failed to remove bridge interface (%s): %v", bridge.DefaultBridgeName, err)
		}
	}
}

func setupInitLayer(idMapping *idtools.IdentityMapping) func(containerfs.ContainerFS) error {
	return func(initPath containerfs.ContainerFS) error {
		return initlayer.Setup(initPath, idMapping.RootPair())
	}
}

// Parse the remapped root (user namespace) option, which can be one of:
//   username            - valid username from /etc/passwd
//   username:groupname  - valid username; valid groupname from /etc/group
//   uid                 - 32-bit unsigned int valid Linux UID value
//   uid:gid             - uid value; 32-bit unsigned int Linux GID value
//
//  If no groupname is specified, and a username is specified, an attempt
//  will be made to lookup a gid for that username as a groupname
//
//  If names are used, they are verified to exist in passwd/group
func parseRemappedRoot(usergrp string) (string, string, error) {

	var (
		userID, groupID     int
		username, groupname string
	)

	idparts := strings.Split(usergrp, ":")
	if len(idparts) > 2 {
		return "", "", fmt.Errorf("Invalid user/group specification in --userns-remap: %q", usergrp)
	}

	if uid, err := strconv.ParseInt(idparts[0], 10, 32); err == nil {
		// must be a uid; take it as valid
		userID = int(uid)
		luser, err := idtools.LookupUID(userID)
		if err != nil {
			return "", "", fmt.Errorf("Uid %d has no entry in /etc/passwd: %v", userID, err)
		}
		username = luser.Name
		if len(idparts) == 1 {
			// if the uid was numeric and no gid was specified, take the uid as the gid
			groupID = userID
			lgrp, err := idtools.LookupGID(groupID)
			if err != nil {
				return "", "", fmt.Errorf("Gid %d has no entry in /etc/group: %v", groupID, err)
			}
			groupname = lgrp.Name
		}
	} else {
		lookupName := idparts[0]
		// special case: if the user specified "default", they want Docker to create or
		// use (after creation) the "dockremap" user/group for root remapping
		if lookupName == defaultIDSpecifier {
			lookupName = defaultRemappedID
		}
		luser, err := idtools.LookupUser(lookupName)
		if err != nil && idparts[0] != defaultIDSpecifier {
			// error if the name requested isn't the special "dockremap" ID
			return "", "", fmt.Errorf("Error during uid lookup for %q: %v", lookupName, err)
		} else if err != nil {
			// special case-- if the username == "default", then we have been asked
			// to create a new entry pair in /etc/{passwd,group} for which the /etc/sub{uid,gid}
			// ranges will be used for the user and group mappings in user namespaced containers
			_, _, err := idtools.AddNamespaceRangesUser(defaultRemappedID)
			if err == nil {
				return defaultRemappedID, defaultRemappedID, nil
			}
			return "", "", fmt.Errorf("Error during %q user creation: %v", defaultRemappedID, err)
		}
		username = luser.Name
		if len(idparts) == 1 {
			// we only have a string username, and no group specified; look up gid from username as group
			group, err := idtools.LookupGroup(lookupName)
			if err != nil {
				return "", "", fmt.Errorf("Error during gid lookup for %q: %v", lookupName, err)
			}
			groupname = group.Name
		}
	}

	if len(idparts) == 2 {
		// groupname or gid is separately specified and must be resolved
		// to an unsigned 32-bit gid
		if gid, err := strconv.ParseInt(idparts[1], 10, 32); err == nil {
			// must be a gid, take it as valid
			groupID = int(gid)
			lgrp, err := idtools.LookupGID(groupID)
			if err != nil {
				return "", "", fmt.Errorf("Gid %d has no entry in /etc/passwd: %v", groupID, err)
			}
			groupname = lgrp.Name
		} else {
			// not a number; attempt a lookup
			if _, err := idtools.LookupGroup(idparts[1]); err != nil {
				return "", "", fmt.Errorf("Error during groupname lookup for %q: %v", idparts[1], err)
			}
			groupname = idparts[1]
		}
	}
	return username, groupname, nil
}

func setupRemappedRoot(config *config.Config) (*idtools.IdentityMapping, error) {
	if runtime.GOOS != "linux" && config.RemappedRoot != "" {
		return nil, fmt.Errorf("User namespaces are only supported on Linux")
	}

	// if the daemon was started with remapped root option, parse
	// the config option to the int uid,gid values
	if config.RemappedRoot != "" {
		username, groupname, err := parseRemappedRoot(config.RemappedRoot)
		if err != nil {
			return nil, err
		}
		if username == "root" {
			// Cannot setup user namespaces with a 1-to-1 mapping; "--root=0:0" is a no-op
			// effectively
			logrus.Warn("User namespaces: root cannot be remapped with itself; user namespaces are OFF")
			return &idtools.IdentityMapping{}, nil
		}
		logrus.Infof("User namespaces: ID ranges will be mapped to subuid/subgid ranges of: %s:%s", username, groupname)
		// update remapped root setting now that we have resolved them to actual names
		config.RemappedRoot = fmt.Sprintf("%s:%s", username, groupname)

		mappings, err := idtools.NewIdentityMapping(username, groupname)
		if err != nil {
			return nil, errors.Wrap(err, "Can't create ID mappings")
		}
		return mappings, nil
	}
	return &idtools.IdentityMapping{}, nil
}

func setupDaemonRoot(config *config.Config, rootDir string, rootIdentity idtools.Identity) error {
	config.Root = rootDir
	// the docker root metadata directory needs to have execute permissions for all users (g+x,o+x)
	// so that syscalls executing as non-root, operating on subdirectories of the graph root
	// (e.g. mounted layers of a container) can traverse this path.
	// The user namespace support will create subdirectories for the remapped root host uid:gid
	// pair owned by that same uid:gid pair for proper write access to those needed metadata and
	// layer content subtrees.
	if _, err := os.Stat(rootDir); err == nil {
		// root current exists; verify the access bits are correct by setting them
		if err = os.Chmod(rootDir, 0711); err != nil {
			return err
		}
	} else if os.IsNotExist(err) {
		// no root exists yet, create it 0711 with root:root ownership
		if err := os.MkdirAll(rootDir, 0711); err != nil {
			return err
		}
	}

	// if user namespaces are enabled we will create a subtree underneath the specified root
	// with any/all specified remapped root uid/gid options on the daemon creating
	// a new subdirectory with ownership set to the remapped uid/gid (so as to allow
	// `chdir()` to work for containers namespaced to that uid/gid)
	if config.RemappedRoot != "" {
		config.Root = filepath.Join(rootDir, fmt.Sprintf("%d.%d", rootIdentity.UID, rootIdentity.GID))
		logrus.Debugf("Creating user namespaced daemon root: %s", config.Root)
		// Create the root directory if it doesn't exist
		if err := idtools.MkdirAllAndChown(config.Root, 0700, rootIdentity); err != nil {
			return fmt.Errorf("Cannot create daemon root: %s: %v", config.Root, err)
		}
		// we also need to verify that any pre-existing directories in the path to
		// the graphroot won't block access to remapped root--if any pre-existing directory
		// has strict permissions that don't allow "x", container start will fail, so
		// better to warn and fail now
		dirPath := config.Root
		for {
			dirPath = filepath.Dir(dirPath)
			if dirPath == "/" {
				break
			}
			if !idtools.CanAccess(dirPath, rootIdentity) {
				return fmt.Errorf("a subdirectory in your graphroot path (%s) restricts access to the remapped root uid/gid; please fix by allowing 'o+x' permissions on existing directories", config.Root)
			}
		}
	}

	if err := setupDaemonRootPropagation(config); err != nil {
		logrus.WithError(err).WithField("dir", config.Root).Warn("Error while setting daemon root propagation, this is not generally critical but may cause some functionality to not work or fallback to less desirable behavior")
	}
	return nil
}

func setupDaemonRootPropagation(cfg *config.Config) error {
	rootParentMount, options, err := getSourceMount(cfg.Root)
	if err != nil {
		return errors.Wrap(err, "error getting daemon root's parent mount")
	}

	var cleanupOldFile bool
	cleanupFile := getUnmountOnShutdownPath(cfg)
	defer func() {
		if !cleanupOldFile {
			return
		}
		if err := os.Remove(cleanupFile); err != nil && !os.IsNotExist(err) {
			logrus.WithError(err).WithField("file", cleanupFile).Warn("could not clean up old root propagation unmount file")
		}
	}()

	if hasMountinfoOption(options, sharedPropagationOption, slavePropagationOption) {
		cleanupOldFile = true
		return nil
	}

	if err := mount.MakeShared(cfg.Root); err != nil {
		return errors.Wrap(err, "could not setup daemon root propagation to shared")
	}

	// check the case where this may have already been a mount to itself.
	// If so then the daemon only performed a remount and should not try to unmount this later.
	if rootParentMount == cfg.Root {
		cleanupOldFile = true
		return nil
	}

	if err := ioutil.WriteFile(cleanupFile, nil, 0600); err != nil {
		return errors.Wrap(err, "error writing file to signal mount cleanup on shutdown")
	}
	return nil
}

// getUnmountOnShutdownPath generates the path to used when writing the file that signals to the daemon that on shutdown
// the daemon root should be unmounted.
func getUnmountOnShutdownPath(config *config.Config) string {
	return filepath.Join(config.ExecRoot, "unmount-on-shutdown")
}

// registerLinks writes the links to a file.
func (daemon *Daemon) registerLinks(container *container.Container, hostConfig *containertypes.HostConfig) error {
	if hostConfig == nil || hostConfig.NetworkMode.IsUserDefined() {
		return nil
	}

	for _, l := range hostConfig.Links {
		name, alias, err := opts.ParseLink(l)
		if err != nil {
			return err
		}
		child, err := daemon.GetContainer(name)
		if err != nil {
			return errors.Wrapf(err, "could not get container for %s", name)
		}
		for child.HostConfig.NetworkMode.IsContainer() {
			parts := strings.SplitN(string(child.HostConfig.NetworkMode), ":", 2)
			child, err = daemon.GetContainer(parts[1])
			if err != nil {
				return errors.Wrapf(err, "Could not get container for %s", parts[1])
			}
		}
		if child.HostConfig.NetworkMode.IsHost() {
			return runconfig.ErrConflictHostNetworkAndLinks
		}
		if err := daemon.registerLink(container, child, alias); err != nil {
			return err
		}
	}

	// After we load all the links into the daemon
	// set them to nil on the hostconfig
	_, err := container.WriteHostConfig()
	return err
}

// conditionalMountOnStart is a platform specific helper function during the
// container start to call mount.
func (daemon *Daemon) conditionalMountOnStart(container *container.Container) error {
	return daemon.Mount(container)
}

// conditionalUnmountOnCleanup is a platform specific helper function called
// during the cleanup of a container to unmount.
func (daemon *Daemon) conditionalUnmountOnCleanup(container *container.Container) error {
	return daemon.Unmount(container)
}

func copyBlkioEntry(entries []*containerd_cgroups.BlkIOEntry) []types.BlkioStatEntry {
	out := make([]types.BlkioStatEntry, len(entries))
	for i, re := range entries {
		out[i] = types.BlkioStatEntry{
			Major: re.Major,
			Minor: re.Minor,
			Op:    re.Op,
			Value: re.Value,
		}
	}
	return out
}

func (daemon *Daemon) stats(c *container.Container) (*types.StatsJSON, error) {
	if !c.IsRunning() {
		return nil, errNotRunning(c.ID)
	}
	cs, err := daemon.containerd.Stats(context.Background(), c.ID)
	if err != nil {
		if strings.Contains(err.Error(), "container not found") {
			return nil, containerNotFound(c.ID)
		}
		return nil, err
	}
	s := &types.StatsJSON{}
	s.Read = cs.Read
	stats := cs.Metrics
	if stats.Blkio != nil {
		s.BlkioStats = types.BlkioStats{
			IoServiceBytesRecursive: copyBlkioEntry(stats.Blkio.IoServiceBytesRecursive),
			IoServicedRecursive:     copyBlkioEntry(stats.Blkio.IoServicedRecursive),
			IoQueuedRecursive:       copyBlkioEntry(stats.Blkio.IoQueuedRecursive),
			IoServiceTimeRecursive:  copyBlkioEntry(stats.Blkio.IoServiceTimeRecursive),
			IoWaitTimeRecursive:     copyBlkioEntry(stats.Blkio.IoWaitTimeRecursive),
			IoMergedRecursive:       copyBlkioEntry(stats.Blkio.IoMergedRecursive),
			IoTimeRecursive:         copyBlkioEntry(stats.Blkio.IoTimeRecursive),
			SectorsRecursive:        copyBlkioEntry(stats.Blkio.SectorsRecursive),
		}
	}
	if stats.CPU != nil {
		s.CPUStats = types.CPUStats{
			CPUUsage: types.CPUUsage{
				TotalUsage:        stats.CPU.Usage.Total,
				PercpuUsage:       stats.CPU.Usage.PerCPU,
				UsageInKernelmode: stats.CPU.Usage.Kernel,
				UsageInUsermode:   stats.CPU.Usage.User,
			},
			ThrottlingData: types.ThrottlingData{
				Periods:          stats.CPU.Throttling.Periods,
				ThrottledPeriods: stats.CPU.Throttling.ThrottledPeriods,
				ThrottledTime:    stats.CPU.Throttling.ThrottledTime,
			},
		}
	}

	if stats.Memory != nil {
		raw := make(map[string]uint64)
		raw["cache"] = stats.Memory.Cache
		raw["rss"] = stats.Memory.RSS
		raw["rss_huge"] = stats.Memory.RSSHuge
		raw["mapped_file"] = stats.Memory.MappedFile
		raw["dirty"] = stats.Memory.Dirty
		raw["writeback"] = stats.Memory.Writeback
		raw["pgpgin"] = stats.Memory.PgPgIn
		raw["pgpgout"] = stats.Memory.PgPgOut
		raw["pgfault"] = stats.Memory.PgFault
		raw["pgmajfault"] = stats.Memory.PgMajFault
		raw["inactive_anon"] = stats.Memory.InactiveAnon
		raw["active_anon"] = stats.Memory.ActiveAnon
		raw["inactive_file"] = stats.Memory.InactiveFile
		raw["active_file"] = stats.Memory.ActiveFile
		raw["unevictable"] = stats.Memory.Unevictable
		raw["hierarchical_memory_limit"] = stats.Memory.HierarchicalMemoryLimit
		raw["hierarchical_memsw_limit"] = stats.Memory.HierarchicalSwapLimit
		raw["total_cache"] = stats.Memory.TotalCache
		raw["total_rss"] = stats.Memory.TotalRSS
		raw["total_rss_huge"] = stats.Memory.TotalRSSHuge
		raw["total_mapped_file"] = stats.Memory.TotalMappedFile
		raw["total_dirty"] = stats.Memory.TotalDirty
		raw["total_writeback"] = stats.Memory.TotalWriteback
		raw["total_pgpgin"] = stats.Memory.TotalPgPgIn
		raw["total_pgpgout"] = stats.Memory.TotalPgPgOut
		raw["total_pgfault"] = stats.Memory.TotalPgFault
		raw["total_pgmajfault"] = stats.Memory.TotalPgMajFault
		raw["total_inactive_anon"] = stats.Memory.TotalInactiveAnon
		raw["total_active_anon"] = stats.Memory.TotalActiveAnon
		raw["total_inactive_file"] = stats.Memory.TotalInactiveFile
		raw["total_active_file"] = stats.Memory.TotalActiveFile
		raw["total_unevictable"] = stats.Memory.TotalUnevictable

		if stats.Memory.Usage != nil {
			s.MemoryStats = types.MemoryStats{
				Stats:    raw,
				Usage:    stats.Memory.Usage.Usage,
				MaxUsage: stats.Memory.Usage.Max,
				Limit:    stats.Memory.Usage.Limit,
				Failcnt:  stats.Memory.Usage.Failcnt,
			}
		} else {
			s.MemoryStats = types.MemoryStats{
				Stats: raw,
			}
		}

		// if the container does not set memory limit, use the machineMemory
		if s.MemoryStats.Limit > daemon.machineMemory && daemon.machineMemory > 0 {
			s.MemoryStats.Limit = daemon.machineMemory
		}
	}

	if stats.Pids != nil {
		s.PidsStats = types.PidsStats{
			Current: stats.Pids.Current,
			Limit:   stats.Pids.Limit,
		}
	}

	return s, nil
}

// setDefaultIsolation determines the default isolation mode for the
// daemon to run in. This is only applicable on Windows
func (daemon *Daemon) setDefaultIsolation() error {
	return nil
}

// setupDaemonProcess sets various settings for the daemon's process
func setupDaemonProcess(config *config.Config) error {
	// setup the daemons oom_score_adj
	if err := setupOOMScoreAdj(config.OOMScoreAdjust); err != nil {
		return err
	}
	if err := setMayDetachMounts(); err != nil {
		logrus.WithError(err).Warn("Could not set may_detach_mounts kernel parameter")
	}
	return nil
}

// This is used to allow removal of mountpoints that may be mounted in other
// namespaces on RHEL based kernels starting from RHEL 7.4.
// Without this setting, removals on these RHEL based kernels may fail with
// "device or resource busy".
// This setting is not available in upstream kernels as it is not configurable,
// but has been in the upstream kernels since 3.15.
func setMayDetachMounts() error {
	f, err := os.OpenFile("/proc/sys/fs/may_detach_mounts", os.O_WRONLY, 0)
	if err != nil {
		if os.IsNotExist(err) {
			return nil
		}
		return errors.Wrap(err, "error opening may_detach_mounts kernel config file")
	}
	defer f.Close()

	_, err = f.WriteString("1")
	if os.IsPermission(err) {
		// Setting may_detach_mounts does not work in an
		// unprivileged container. Ignore the error, but log
		// it if we appear not to be in that situation.
		if !rsystem.RunningInUserNS() {
			logrus.Debugf("Permission denied writing %q to /proc/sys/fs/may_detach_mounts", "1")
		}
		return nil
	}
	return err
}

func setupOOMScoreAdj(score int) error {
	f, err := os.OpenFile("/proc/self/oom_score_adj", os.O_WRONLY, 0)
	if err != nil {
		return err
	}
	defer f.Close()
	stringScore := strconv.Itoa(score)
	_, err = f.WriteString(stringScore)
	if os.IsPermission(err) {
		// Setting oom_score_adj does not work in an
		// unprivileged container. Ignore the error, but log
		// it if we appear not to be in that situation.
		if !rsystem.RunningInUserNS() {
			logrus.Debugf("Permission denied writing %q to /proc/self/oom_score_adj", stringScore)
		}
		return nil
	}

	return err
}

func (daemon *Daemon) initCgroupsPath(path string) error {
	if path == "/" || path == "." {
		return nil
	}

	if daemon.configStore.CPURealtimePeriod == 0 && daemon.configStore.CPURealtimeRuntime == 0 {
		return nil
	}

	// Recursively create cgroup to ensure that the system and all parent cgroups have values set
	// for the period and runtime as this limits what the children can be set to.
	daemon.initCgroupsPath(filepath.Dir(path))

	mnt, root, err := cgroups.FindCgroupMountpointAndRoot("", "cpu")
	if err != nil {
		return err
	}
	// When docker is run inside docker, the root is based of the host cgroup.
	// Should this be handled in runc/libcontainer/cgroups ?
	if strings.HasPrefix(root, "/docker/") {
		root = "/"
	}

	path = filepath.Join(mnt, root, path)
	sysinfo := sysinfo.New(true)
	if err := maybeCreateCPURealTimeFile(sysinfo.CPURealtimePeriod, daemon.configStore.CPURealtimePeriod, "cpu.rt_period_us", path); err != nil {
		return err
	}
	return maybeCreateCPURealTimeFile(sysinfo.CPURealtimeRuntime, daemon.configStore.CPURealtimeRuntime, "cpu.rt_runtime_us", path)
}

func maybeCreateCPURealTimeFile(sysinfoPresent bool, configValue int64, file string, path string) error {
	if sysinfoPresent && configValue != 0 {
		if err := os.MkdirAll(path, 0755); err != nil {
			return err
		}
		if err := ioutil.WriteFile(filepath.Join(path, file), []byte(strconv.FormatInt(configValue, 10)), 0700); err != nil {
			return err
		}
	}
	return nil
}

func (daemon *Daemon) setupSeccompProfile() error {
	if daemon.configStore.SeccompProfile != "" {
		daemon.seccompProfilePath = daemon.configStore.SeccompProfile
		b, err := ioutil.ReadFile(daemon.configStore.SeccompProfile)
		if err != nil {
			return fmt.Errorf("opening seccomp profile (%s) failed: %v", daemon.configStore.SeccompProfile, err)
		}
		daemon.seccompProfile = b
	}
	return nil
}