File: procfs.c

package info (click to toggle)
gridengine 6.2-4
  • links: PTS, VCS
  • area: main
  • in suites: lenny
  • size: 51,532 kB
  • ctags: 51,172
  • sloc: ansic: 418,155; java: 37,080; sh: 22,593; jsp: 7,699; makefile: 5,292; csh: 4,244; xml: 2,901; cpp: 2,086; perl: 1,895; tcl: 1,188; lisp: 669; ruby: 642; yacc: 393; lex: 266
file content (808 lines) | stat: -rw-r--r-- 22,694 bytes parent folder | download
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
/*___INFO__MARK_BEGIN__*/
/*************************************************************************
 * 
 *  The Contents of this file are made available subject to the terms of
 *  the Sun Industry Standards Source License Version 1.2
 * 
 *  Sun Microsystems Inc., March, 2001
 * 
 * 
 *  Sun Industry Standards Source License Version 1.2
 *  =================================================
 *  The contents of this file are subject to the Sun Industry Standards
 *  Source License Version 1.2 (the "License"); You may not use this file
 *  except in compliance with the License. You may obtain a copy of the
 *  License at http://gridengine.sunsource.net/Gridengine_SISSL_license.html
 * 
 *  Software provided under this License is provided on an "AS IS" basis,
 *  WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING,
 *  WITHOUT LIMITATION, WARRANTIES THAT THE SOFTWARE IS FREE OF DEFECTS,
 *  MERCHANTABLE, FIT FOR A PARTICULAR PURPOSE, OR NON-INFRINGING.
 *  See the License for the specific provisions governing your rights and
 *  obligations concerning the Software.
 * 
 *   The Initial Developer of the Original Code is: Sun Microsystems, Inc.
 * 
 *   Copyright: 2001 by Sun Microsystems, Inc.
 * 
 *   All Rights Reserved.
 * 
 ************************************************************************/
/*___INFO__MARK_END__*/
#if !defined(COMPILE_DC)

int verydummyprocfs;

#else

#include <stdio.h>
#include <fcntl.h>
#include <time.h>
#include <errno.h>
#include <sys/ioctl.h>
#include <sys/types.h>
#include <sys/signal.h>

#if !defined(CRAY) && !defined(NECSX4) && !defined(NECSX5) && !defined(AIX)
#include <sys/syscall.h>
#endif

#include <unistd.h>
#include <sys/times.h>
#include <sys/wait.h>
#if defined(FREEBSD) || defined(DARWIN)
#include <sys/time.h>
#endif
#include <sys/resource.h>
#include <dirent.h>
#include <stdlib.h>
#include <string.h>
#include <signal.h>

#if defined(ALPHA) 
#  include <sys/user.h>
#  include <sys/table.h>   
#  include <sys/procfs.h>   
#endif

#if defined(SOLARIS) 
#  include <sys/procfs.h>   
#endif

#if defined(LINUX)
#include <sys/param.h>          /* for HZ (jiffies -> seconds ) */
#endif

#include "uti/sge_stdio.h"
#include "uti/sge_unistd.h"

#include "sge_log.h"
#include "msg_sge.h"
#include "sgermon.h"
#include "basis_types.h"
#include "sgedefs.h"
#include "exec_ifm.h"
#include "pdc.h"

#if !defined(CRAY)
#include "procfs.h"
#endif

#if defined(LINUX) || defined(ALPHA) || defined(SOLARIS)
#   if defined(SVR3)
#      define PROC_DIR "/debug"
#   else
#      define PROC_DIR "/proc"
#   endif
#endif

#if defined(LINUX)

#define BIGLINE 1024

typedef struct _tLinProcStat {     /* process Linux /proc/pid/stat structure */
   int   pr_pid;                  /*  1 process id */
   char  pr_cmd[16];              /*  2 command name */
   char  pr_stat;                 /*  3 process status */
   int   pr_ppid;                 /*  4 parent process id */
   int   pr_pgrp;                 /*  5 process group */
   int   pr_sid;                  /*  6 session id */
   int   pr_tty;                  /*  7 tty line MAJOR << ? + MINOR */
   int   pr_tty_pgrp;             /*  8 tty process group */
   long  pr_flags;                /*  9 process flags PF_* in <linux/sched.h> */
   long  pr_min_flt;              /* 10 minor page faults */
   long  pr_cmin_flt;             /* 11 children minor page faults */
   long  pr_maj_flt;              /* 12 major page faults */
   long  pr_cmaj_flt;             /* 13 children major page faults */
   long  pr_utime;                /* 14 user time */
   long  pr_stime;                /* 15 system time */
   long  pr_cutime;               /* 16 children user time */
   long  pr_cstime;               /* 17 children system time */
   long  pr_counter;		        /* 18 jiffies */
   long  pr_pri;                  /* 19 priority (nice) */
   long  pr_tmout;                /* 20 Timeout time for scheduling */
   long  pr_it_real_value;        /* 21 itimer real value */
   long  pr_start;                /* 22 start of execution in jiffies since boot*/
   long  pr_vsize;                /* 23 total size t + d + s NOT pages */
   long  pr_rss;                  /* 24 resident set size pages */
   long  pr_rlim_cur;             /* 25 current rlimit ro rss */
   long  pr_start_code;           /* 26 start of code */
   long  pr_end_code;             /* 27 end of code */
   long  pr_start_stack;          /* 28 start of stack */
   long  pr_esp;                  /* 29 head of stack (stack pointer) */
   long  pr_eip;                  /* 30 instruction pointer */
   long  pr_signal;               /* 31 pending signals mask */
   long  pr_blocked;              /* 32 blocked signals mask */
   long  pr_sigignore;            /* 33 ignored signals mask */
   long  pr_sigcatch;             /* 34 catched signals mask */
   long  pr_wchan;                /* 35 WCHAN (seems to be a return address) */
} tLinProcStat;
#endif


/*-----------------------------------------------------------------------*/
#if defined(LINUX) || defined(ALPHA) || defined(SOLARIS)

static DIR *cwd;
static struct dirent *dent;

#if defined(LINUX)

int groups_in_proc (void) 
{
   char buf[1024];
   FILE* fd = (FILE*) NULL;
   
   if (!(fd = fopen(PROC_DIR "/self/status", "r"))) {
      return 0;
   }
   while (fgets(buf, sizeof(buf), fd)) {
      if (strcmp("Groups:", strtok(buf, "\t"))==0) {
         FCLOSE(fd);
         return 1;
      }
   }
   FCLOSE(fd);
   return 0;
FCLOSE_ERROR:
   return 0;
}

#endif


/* search in job list for the pid
   return the proc element */
static lnk_link_t *find_pid_in_jobs(pid_t pid, lnk_link_t *job_list)
{
   lnk_link_t *job, *proc = NULL;
   proc_elem_t *proc_elem = NULL;
   job_elem_t *job_elem = NULL;

   /*
    * try to find a matching job
    */
   for (job=job_list->next; job != job_list; job=job->next) {

      job_elem = LNK_DATA(job, job_elem_t, link);

      /*
       * try to find process in this jobs' proc list
       */

      for (proc=job_elem->procs.next; proc != &job_elem->procs;
               proc=proc->next) {

         proc_elem = LNK_DATA(proc, proc_elem_t, link);
         if (proc_elem->proc.pd_pid == pid)
            break; /* found it */
      }

      if (proc == &job_elem->procs) {
         /* end of procs list - no process found - try next job */
         proc = NULL;
      } else
         /* found a process */
         break;
   }

   return proc;
}


static void touch_time_stamp(const char *d_name, int time_stamp, lnk_link_t *job_list)
{
   pid_t pid;
   proc_elem_t *proc_elem;
   lnk_link_t *proc;

   DENTER(TOP_LAYER, "touch_time_stamp");

   sscanf(d_name, pid_t_fmt, &pid);
   if ((proc = find_pid_in_jobs(pid, job_list))) {
      proc_elem = LNK_DATA(proc, proc_elem_t, link);
      proc_elem->proc.pd_tstamp = time_stamp;
#ifdef MONITOR_PDC
      INFO((SGE_EVENT, "found job to process %s: set time stamp\n", d_name));
#endif
   }
#ifdef MONITOR_PDC
   else
      INFO((SGE_EVENT, "found no job to process %s\n", d_name));
#endif

   DEXIT;
   return;
}

void procfs_kill_addgrpid(gid_t add_grp_id, int sig,
   tShepherd_trace shepherd_trace)
{
   char procnam[128];
   int i;
   int groups=0;
   u_long32 max_groups;
   gid_t *list;
#if defined(SOLARIS) || defined(ALPHA)
   int fd;
   prcred_t proc_cred;
#elif defined(LINUX)
   FILE *fp;
   char buffer[1024];
   uid_t uids[4];
   gid_t gids[4];
#endif

   DENTER(TOP_LAYER, "procfs_kill_addgrpid");

   /* quick return in case of invalid add. group id */
   if (add_grp_id == 0) {
      DEXIT;
      return;
   }

   max_groups = sge_sysconf(SGE_SYSCONF_NGROUPS_MAX);
   if (max_groups <= 0)
      if (shepherd_trace) {
         char err_str[256];

         sprintf(err_str, MSG_SGE_NGROUPS_MAXOSRECONFIGURATIONNECESSARY );
         shepherd_trace(err_str);
      }
/*
 * INSURE detects a WRITE_OVERFLOW when getgroups was invoked (LINUX).
 * Is this a bug in the kernel or in INSURE?
 */
#if defined(LINUX)
   list = (gid_t*) malloc(2*max_groups*sizeof(gid_t));
#else
   list = (gid_t*) malloc(max_groups*sizeof(gid_t));
#endif
   if (list == NULL)
      if (shepherd_trace) {
         char err_str[256];

         sprintf(err_str, MSG_SGE_PROCFSKILLADDGRPIDMALLOCFAILED );
         shepherd_trace(err_str);
      }

   pt_open();

   /* find next valid entry in procfs  */
   while ((dent = readdir(cwd))) {
      if (!dent->d_name)
         continue;
      if (!dent->d_name[0])
         continue;

      if (!strcmp(dent->d_name, "..") || !strcmp(dent->d_name, "."))
         continue;

      if (atoi(dent->d_name) == 0)
         continue;

#if defined(SOLARIS) || defined(ALPHA)
      sprintf(procnam, "%s/%s", PROC_DIR, dent->d_name);
      if ((fd = open(procnam, O_RDONLY, 0)) == -1) {
         DPRINTF(("open(%s) failed: %s\n", procnam, strerror(errno)));
         continue;
      }
#elif defined(LINUX)
      if (!strcmp(dent->d_name, "self"))
         continue;

      sprintf(procnam, "%s/%s/status", PROC_DIR, dent->d_name);
      if (!(fp = fopen(procnam, "r")))
         continue;
#endif

#if defined(SOLARIS) || defined(ALPHA)
      /* get number of groups */
      if (ioctl(fd, PIOCCRED, &proc_cred) == -1) {
         close(fd);
         continue;
      }

      /* get list of supplementary groups */
      groups = proc_cred.pr_ngroups;
      if (ioctl(fd, PIOCGROUPS, list) == -1) {
         close(fd);
         continue;
      }

#elif defined(LINUX)
      /* get number of groups and current uids, gids
       * uids[0], gids[0] => UID and GID
       * uids[1], gids[1] => EUID and EGID
       * uids[2], gids[2] => SUID and SGID
       * uids[3], gids[3] => FSUID and FSGID
       */
      groups = 0;
      while (fgets(buffer, sizeof(buffer), fp)) {
         char *label = NULL;
         char *token = NULL;

         label = strtok(buffer, " \t\n");
         if (label) {
            if (!strcmp("Groups:", label)) {
               while ((token = strtok((char*) NULL, " \t\n"))) {
                  list[groups]=(gid_t) atol(token);
                  groups++;
               }
            } else if (!strcmp("Uid:", label)) {
               int i = 0;

               while ((i < 4) && (token = strtok((char*) NULL, " \t\n"))) {
                  uids[i]=(uid_t) atol(token);
                  i++;
               }
            } else if (!strcmp("Gid:", label)) {
               int i = 0;

               while ((i < 4) && (token = strtok((char*) NULL, " \t\n"))) {
                  gids[i]=(gid_t) atol(token);
                  i++;
               }
            }
         }
      }
#endif

#if defined(SOLARIS) || defined(ALPHA)
      close(fd);
#elif defined(LINUX)
      FCLOSE(fp);
FCLOSE_ERROR:
#endif

      /* send each process a signal which belongs to add_grg_id */
      for (i = 0; i < groups; i++) {
         if (list[i] == add_grp_id) {
            pid_t pid;
            pid = (pid_t) atol(dent->d_name);

#if defined(LINUX)
            /* if UID, GID, EUID and EGID == 0
             *  don't kill the process!!! - it could be the rpc.nfs-deamon
             */
            if (!(uids[0] == 0 && gids[0] == 0 &&
                  uids[1] == 0 && gids[1] == 0)) {
#elif defined(SOLARIS) || defined(ALPHA)
            if (!(proc_cred.pr_ruid == 0 && proc_cred.pr_rgid == 0 &&
                  proc_cred.pr_euid == 0 && proc_cred.pr_egid == 0)) {
#endif

               if (shepherd_trace) {
                  char err_str[256];

                  sprintf(err_str, MSG_SGE_KILLINGPIDXY_UI , sge_u32c(pid), groups);
                  shepherd_trace(err_str);
               }

               kill(pid, sig);

            } else {
               if (shepherd_trace) {
                  char err_str[256];

                  sprintf(err_str, MSG_SGE_DONOTKILLROOTPROCESSXY_UI ,
                     sge_u32c(atol(dent->d_name)), groups);
                  shepherd_trace(err_str);
               }
            }

            break;
         }
      }
   }
   pt_close();
   free(list);
   DEXIT;
}

int pt_open(void)
{
   cwd = opendir(PROC_DIR);
   return !cwd;
}
void pt_close(void)
{
   closedir(cwd);
}

int pt_dispatch_proc_to_job(
lnk_link_t *job_list,
int time_stamp 
) {
   char procnam[128];
   int fd = -1;
#if defined(LINUX)
   char buffer[BIGLINE];
   tLinProcStat pr;
   SGE_STRUCT_STAT fst;
#else
   prstatus_t pr;
   prpsinfo_t pri;
#endif

#if defined(SOLARIS) || defined(ALPHA)   
   prcred_t proc_cred;
#endif

   int ret;
   u_long32 max_groups;
   gid_t *list;
   int groups=0;

   proc_elem_t *proc_elem = NULL;
   job_elem_t *job_elem = NULL;
   lnk_link_t *curr;
   double old_time = 0;
   uint64 old_vmem = 0;

   DENTER(TOP_LAYER, "pt_dispatch_proc_to_job");

   max_groups = sge_sysconf(SGE_SYSCONF_NGROUPS_MAX);
   if (max_groups <= 0) {
      ERROR((SGE_EVENT, MSG_SGE_NGROUPS_MAXOSRECONFIGURATIONNECESSARY));
      DEXIT;
      return 1;  
   }   

   list = (gid_t*) malloc(max_groups*sizeof(gid_t));
   if (list == NULL) {
      ERROR((SGE_EVENT, MSG_SGE_PTDISPATCHPROCTOJOBMALLOCFAILED));
      DEXIT;
      return 1;
   }

   /* find next valid entry in procfs */ 
   while ((dent = readdir(cwd))) {
      char *pidname;

      if (!dent->d_name)
         continue;
      if (!dent->d_name[0])
         continue;

      if (!strcmp(dent->d_name, "..") || !strcmp(dent->d_name, "."))
         continue;

      if (dent->d_name[0] == '.')
          pidname = &dent->d_name[1];
      else
          pidname = dent->d_name;

      if (atoi(pidname) == 0)
         continue;

#if defined(LINUX)
      sprintf(procnam, "%s/%s/stat", PROC_DIR, dent->d_name);
#else
      sprintf(procnam, "%s/%s", PROC_DIR, dent->d_name);
#endif
      if ((fd = open(procnam, O_RDONLY, 0)) == -1) {
         if (errno != ENOENT) {
#ifdef MONITOR_PDC
            if (errno == EACCES)
               INFO((SGE_EVENT, "(uid:"gid_t_fmt" euid:"gid_t_fmt") could not open %s: %s\n",
                        getuid(), geteuid(), procnam, strerror(errno)));
            else
               INFO((SGE_EVENT, "could not open %s: %s\n", procnam, strerror(errno)));
#endif
            touch_time_stamp(dent->d_name, time_stamp, job_list);
         }
         continue;
      }

      /** 
       ** get a list of supplementary group ids to decide
       ** whether this process will be needed;
       ** read also prstatus
       **/

#  if defined(LINUX)

      /* 
       * Read the line and append a 0-Byte 
       */
      if ((ret = read(fd, buffer, BIGLINE-1))<=0) {
         close(fd);
         if (ret == -1 && errno != ENOENT) {
#ifdef MONITOR_PDC
            INFO((SGE_EVENT, "could not read %s: %s\n", procnam, strerror(errno)));
#endif
            touch_time_stamp(dent->d_name, time_stamp, job_list);
         }
         continue;
      }
      buffer[BIGLINE-1] = '\0';
      
      if (SGE_FSTAT(fd, &fst)) {
         close(fd);
         if (errno != ENOENT) {
#ifdef MONITOR_PDC
            INFO((SGE_EVENT, "could not fstat %s: %s\n", procnam, strerror(errno)));
#endif
            touch_time_stamp(dent->d_name, time_stamp, job_list);
         }
         continue;
      }

      /* 
       * get prstatus
       */
      ret = sscanf(buffer, 
                   "%d %s %c %d %d %d %d %d %lu %lu \
                    %lu %lu %lu %ld %ld %ld %ld %ld %ld %lu \
                    %lu %ld %lu %lu %lu %lu %lu %lu %lu %lu \
                    %lu %lu %lu %lu %lu",
                    &pr.pr_pid,
                    pr.pr_cmd, 
                    &pr.pr_stat,
                    &pr.pr_ppid,
                    &pr.pr_pgrp,
                    &pr.pr_sid,
                    &pr.pr_tty,
                    &pr.pr_tty_pgrp,
                    &pr.pr_flags,   
                    &pr.pr_min_flt, 
                    &pr.pr_cmin_flt,
                    &pr.pr_maj_flt, 
                    &pr.pr_cmaj_flt,
                    &pr.pr_utime,   
                    &pr.pr_stime,   
                    &pr.pr_cutime,  
                    &pr.pr_cstime,
                    &pr.pr_counter,  
                    &pr.pr_pri,     
                    &pr.pr_tmout,   
                    &pr.pr_it_real_value,
                    &pr.pr_start,
                    &pr.pr_vsize,
                    &pr.pr_rss,  
                    &pr.pr_rlim_cur,
                    &pr.pr_start_code,
                    &pr.pr_end_code,  
                    &pr.pr_start_stack,
                    &pr.pr_esp,
                    &pr.pr_eip,
                    &pr.pr_signal,
                    &pr.pr_blocked,
                    &pr.pr_sigignore,
                    &pr.pr_sigcatch, 
                    &pr.pr_wchan);   

      if (ret != 35) {
         close(fd);
         continue;
      }

      /* 
       * get number of groups; 
       * get list of supplementary groups 
       */
      {
         char procnam[256];
         char buf[1024];
         FILE* f = (FILE*) NULL;
   
         sprintf(procnam,  "%s/%s/status", PROC_DIR, dent->d_name);
         if (!(f = fopen(procnam, "r"))) {
            close(fd);
            continue;
         }
         groups = 0;
         while (fgets(buf, sizeof(buf), f)) {
            if (strcmp("Groups:", strtok(buf, "\t"))==0) {
               char *token;
                  
               while ((token=strtok((char*) NULL, " "))) {
                  list[groups]=atol(token);
                  groups++;
               }
               break;
            }
         }
         FCLOSE(f);
FCLOSE_ERROR:
         ;
      } 
#  elif defined(SOLARIS) || defined(ALPHA)
      
      /* 
       * get prstatus 
       */
      if (ioctl(fd, PIOCSTATUS, &pr)==-1) {
         close(fd);
         if (errno != ENOENT) {
#ifdef MONITOR_PDC
            INFO((SGE_EVENT, "could not ioctl(PIOCSTATUS) %s: %s\n", procnam, strerror(errno)));
#endif
            touch_time_stamp(dent->d_name, time_stamp, job_list);
         }
         continue;
      }
                                    
      /* 
       * get number of groups 
       */
      ret=ioctl(fd, PIOCCRED, &proc_cred);
      if (ret < 0) {
         close(fd);
         if (errno != ENOENT) {
#ifdef MONITOR_PDC
            INFO((SGE_EVENT, "could not ioctl(PIOCCRED) %s: %s\n", procnam, strerror(errno)));
#endif
            touch_time_stamp(dent->d_name, time_stamp, job_list);
         }
         continue;
      }
      
      /* 
       * get list of supplementary groups 
       */
      groups = proc_cred.pr_ngroups;
      ret=ioctl(fd, PIOCGROUPS, list);
      if (ret<0) {
         close(fd);
         if (errno != ENOENT) {
#ifdef MONITOR_PDC
            INFO((SGE_EVENT, "could not ioctl(PIOCCRED) %s: %s\n", procnam, strerror(errno)));
#endif
            touch_time_stamp(dent->d_name, time_stamp, job_list);
         }
         continue;
      }

#  endif

      /* 
       * try to find a matching job 
       */
      for (curr=job_list->next; curr != job_list; curr=curr->next) {
         int found_it = 0;
         int group;
         
         job_elem = LNK_DATA(curr, job_elem_t, link);
         for (group=0; !found_it && group<groups; group++) {
            if (job_elem->job.jd_jid == list[group]) {
               found_it = 1;
            }
         }
         if (found_it)
            break;
      }

      if (curr == job_list) { /* this is not a traced process */ 
         close(fd);
         continue;
      }

      /* we always read only one entry per function call
         the while loop is needed to read next one */
      break;
   } /* while */

   free(list);

   if (!dent) {/* visited all files in procfs */
      DEXIT;
      return 1;
   }
   /* 
    * try to find process in this jobs' proc list 
    */

   for (curr=job_elem->procs.next; curr != &job_elem->procs; 
            curr=curr->next) {
      proc_elem = LNK_DATA(curr, proc_elem_t, link);
      
      if (proc_elem->proc.pd_pid == pr.pr_pid)
         break;
   }

   if (curr == &job_elem->procs) { 
      /* new process, add a proc element into jobs proc list */
      if (!(proc_elem=(proc_elem_t *)malloc(sizeof(proc_elem_t)))) {
         if (fd >= 0)
            close(fd);
         DEXIT;
         return 0;
      }
      memset(proc_elem, 0, sizeof(proc_elem_t));
      proc_elem->proc.pd_length = sizeof(psProc_t);
      proc_elem->proc.pd_state  = 1; /* active */
      LNK_ADD(job_elem->procs.prev, &proc_elem->link);
      job_elem->job.jd_proccount++;

#ifdef MONITOR_PDC
      {
         double utime, stime;
#if defined(LINUX)
         utime = ((double)pr.pr_utime)/HZ; 
         stime = ((double)pr.pr_stime)/HZ;
#else
         utime = pr.pr_utime.tv_sec + pr.pr_utime.tv_nsec*1E-9;
         stime = pr.pr_stime.tv_sec + pr.pr_stime.tv_nsec*1E-9;
#endif
         INFO((SGE_EVENT, "new process "pid_t_fmt" for job "pid_t_fmt" (utime = %f stime = %f)\n", 
               pr.pr_pid, job_elem->job.jd_jid, utime, stime)); 
      }
#endif

   } else {
      /* save previous usage data - needed to build delta usage */
      old_time = proc_elem->proc.pd_utime + proc_elem->proc.pd_stime;
      old_vmem  = proc_elem->vmem;
   }

   proc_elem->proc.pd_tstamp = time_stamp;

   proc_elem->proc.pd_pid    = pr.pr_pid;
#if defined(LINUX)
   proc_elem->proc.pd_utime  = ((double)pr.pr_utime)/HZ;
   proc_elem->proc.pd_stime  = ((double)pr.pr_stime)/HZ;
   /* could retrieve uid/gid using stat() on stat file */
   proc_elem->vmem           = pr.pr_vsize;
#else
   proc_elem->proc.pd_utime  = pr.pr_utime.tv_sec + pr.pr_utime.tv_nsec*1E-9;
   proc_elem->proc.pd_stime  = pr.pr_stime.tv_sec + pr.pr_stime.tv_nsec*1E-9;
    
   /* Don't care if this part fails */
   if (ioctl(fd, PIOCPSINFO, &pri) != -1) {
      proc_elem->proc.pd_uid    = pri.pr_uid;
      proc_elem->proc.pd_gid    = pri.pr_gid;
      proc_elem->vmem           = pri.pr_size * pagesize;
      proc_elem->rss            = pri.pr_rssize * pagesize;
      proc_elem->proc.pd_pstart = pri.pr_start.tv_sec + pri.pr_start.tv_nsec*1E-9;
   }
#endif         

   proc_elem->mem = 
         ((proc_elem->proc.pd_stime + proc_elem->proc.pd_utime) - old_time) * 
         (( old_vmem + proc_elem->vmem)/2);

#if defined(ALPHA)
#define BLOCKSIZE 512
   {
      struct user ua;
      uint64 old_ru_ioblock = proc_elem->ru_ioblock;

      /* need to do a table(2) call for each process to retrieve io usage data */   
      /* get user area stuff */
      if (table(TBL_UAREA, proc_elem->proc.pd_pid, (char *)&ua, 1, sizeof ua) == 1) {
         proc_elem->ru_ioblock = (uint64)(ua.u_ru.ru_inblock + ua.u_ru.ru_oublock);
         proc_elem->delta_chars = (proc_elem->ru_ioblock - old_ru_ioblock)* BLOCKSIZE;
      }
   }
#endif


   close(fd);
   DEXIT;
   return 0;
}
#endif

#endif /* (!COMPILE_DC) */