Stack question

[Padiyath Sreekumaran]

  Hello,
   On one of our Linux server I have the following configuration:
   There are 2 CPU's. cat /proc/meminfo gives the following output:

kumar@:~>  cat /proc/meminfo
MemTotal:      5609156 kB
MemFree:       2693344 kB
Buffers:        307912 kB
Cached:        2191972 kB
SwapCached:          0 kB
Active:         873552 kB
Inactive:      1870992 kB
HighTotal:           0 kB
HighFree:            0 kB
LowTotal:      5609156 kB
LowFree:       2693344 kB
SwapTotal:     2285544 kB
SwapFree:      2285544 kB
Dirty:              32 kB
Writeback:           0 kB
Mapped:         292956 kB
Slab:           145300 kB
Committed_AS:   529164 kB
PageTables:       3600 kB
VmallocTotal: 536870911 kB
VmallocUsed:     18136 kB
VmallocChunk: 536852767 kB
HugePages_Total:     0
HugePages_Free:      0
Hugepagesize:     2048 kB


   limit command gives the following output:
  ============================================
kumar@:~>limit
cputime      unlimited
filesize     unlimited
datasize     unlimited
stacksize    unlimited
coredumpsize    0 kbytes
memoryuse    unlimited
vmemoryuse   unlimited
descriptors  1024
memorylocked    unlimited
maxproc      49152

    When I run a job which takes 2 processors die program crashes
    with the following given error at the beginning.Iam using PVM to run
    parallel jobw. The same job works with 1 processor on the same machine.

    The error is:

 +--------------------------------------------------------------------+
 |                                                                    |
 |                *** INSUFFICIENT MEMORY ALLOCATED ***               |
 |                                                                    |
 | Action required : Increase the integer stack memory size.          |
 |                                                                    |
 | If the situation persists please contact the CFX Customer Helpline |
 | giving the following details:-                                     |
 |                                                                    |
 | Requested space:                  131576 words                     |
 |                                                                    |
 | Current allocated space:        20349540 words                     |
 | Current used space:             20321375 words                     |
 | Current free space:                28165 words                     |
 | Number of free areas:                  3                           |
 |                                                                    |
 +--------------------------------------------------------------------+
 Fatal error generated in gKSelEl
 Message :- FULL : Failed to make a data area for KSelEl
 gKSelEl called by :-  gKVxEl_ZN
--- SIGCHLD (Child exited) @ 0 (0) ---
An error has occurred in cfx5solve:

The CFX-5 solver has terminated without writing a results file.  Command on
host pc4989 exited with return code 0.

It gave the following output:

Contents of /tmp/pvml.1002:

  [t80040000] 08/15 10:27:36 pc4989 (129.129.234.168:32816) LINUX 3.4.4
  [t80040000] 08/15 10:27:36 ready Mon Aug 15 10:27:36 2005

--- SIGCHLD (Child exited) @ 0 (0) ---
--- SIGCHLD (Child exited) @ 0 (0) ---
--- SIGCHLD (Child exited) @ 0 (0) ---
--- SIGCHLD (Child exited) @ 0 (0) ---
[ Process PID=26615 runs in 32 bit mode. ]


     Any help is very much appreciated.

   Regards,
   Kumar
   
------------------------------------------------------------------
Padiyath Sreekumar       |   Tel: +41.56.310.3643
Paul Scherrer Institut   |   email: kumar.padiyath at psi.ch
AIT                      |   Office: WHGA/U132 
WHGA/U132                |   Fax: +41.56.310.3649
CH-5232 Villigen PSI     | 
Switzerland              |
-----------------------------------------------------------------