Warning: Undefined array key "DOKU_PREFS" in /usr/share/dokuwiki/inc/common.php on line 2082
cluster:190 [DokuWiki]

DokuWiki

User Tools

Site Tools

Trace:
cluster:190

Warning: Undefined array key -1 in /usr/share/dokuwiki/inc/html.php on line 1458

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision 		Previous revision
cluster:190 [2020/01/21 14:37]
hmeij07 		cluster:190 [2020/09/28 07:38]
hmeij07
Line 11: Line 11:
  
 This is a replacement for THE BLCR methods we used [[cluster:147|BLCR Checkpoint in OL3 -serial]] or [[cluster:148|BLCR Checkpoint in OL3 - parallel]] ... BLCR is not being developed anymore. Today's brief power outage removed the BLCR kernel module HPCC wide. So learn DMTCP. I have not provided wrappers but you can follow the same logic as we used with BLCR. This is a replacement for THE BLCR methods we used [[cluster:147|BLCR Checkpoint in OL3 -serial]] or [[cluster:148|BLCR Checkpoint in OL3 - parallel]] ... BLCR is not being developed anymore. Today's brief power outage removed the BLCR kernel module HPCC wide. So learn DMTCP. I have not provided wrappers but you can follow the same logic as we used with BLCR.
 +
 +Write your checkpoint files in ''/sanscratch/checkpoints/JOBPID'' so it does not add into your quota.  The scheduler will not create this directory for you, you must do this in your submit job.  Directories will automatically be delete if 120 days old.
  
 <code> <code>
Line 17: Line 19:
  
 # make a directory (first one done by scheduler, second one done by your script) # make a directory (first one done by scheduler, second one done by your script)
-mkdir -p /sanscratch/111 /snascratch/checkpoints/111+mkdir -p /sanscratch/111 /sanscratch/checkpoints/111
 cd /sanscratch/111 cd /sanscratch/111
 +
 +# invoke sample rpogram which generates one line of output
 time ./a.out time ./a.out
  
Line 32: Line 36:
 [hmeij@cottontail2 111]$ cat fid.txt [hmeij@cottontail2 111]$ cat fid.txt
   0.999578495067887       7.978578277139264E-005   0.999578495067887       7.978578277139264E-005
 +
 +# for example, with 24 hours checkpoint interval
 +# launch new coordinator on random port, log port, 24 hour checkpoints 
 +# make sure you create destination dir for checkpoints
 +dmtcp_launch --new-coordinator \
 +  --coord-port 0 --port-file port.txt --interval 86400 \
 +  --ckptdir /sanscratch/checkpoints/111 \
 +  time ./a.out
 +
      
-# run command below with interval 300 (every 5 mins)+# run command above  with interval 300 (every 5 mins) 
 [hmeij@cottontail2 111]$ ps [hmeij@cottontail2 111]$ ps
   PID TTY          TIME CMD   PID TTY          TIME CMD
Line 45: Line 59:
 1 S hmeij    20008     1  0  80   0 -  4665 ep_pol 07:46 pts/1    00:00:00 \\ 1 S hmeij    20008     1  0  80   0 -  4665 ep_pol 07:46 pts/1    00:00:00 \\
 /usr/bin/dmtcp_coordinator --quiet --exit-on-last --daemon /usr/bin/dmtcp_coordinator --quiet --exit-on-last --daemon
 +
 # the random port (in case somebody else is also checkpointing on this host # the random port (in case somebody else is also checkpointing on this host
 [hmeij@cottontail2 111]$ cat port.txt [hmeij@cottontail2 111]$ cat port.txt
Line 55: Line 70:
 -rwxr--r-- 1 hmeij its   12440 Jan 16 13:53 dmtcp_restart_script_24945f6ae3823bbf-40000-fb2d11ea62fa0.sh\\ -rwxr--r-- 1 hmeij its   12440 Jan 16 13:53 dmtcp_restart_script_24945f6ae3823bbf-40000-fb2d11ea62fa0.sh\\
 lrwxrwxrwx 1 hmeij its      60 Jan 16 13:53 dmtcp_restart_script.sh -> dmtcp_restart_script_24945f6ae3823bbf-40000-fb2d11ea62fa0.sh\\ lrwxrwxrwx 1 hmeij its      60 Jan 16 13:53 dmtcp_restart_script.sh -> dmtcp_restart_script_24945f6ae3823bbf-40000-fb2d11ea62fa0.sh\\
 +
 [hmeij@cottontail2 111]$ [hmeij@cottontail2 111]$
 real    63m57.287s real    63m57.287s
Line 65: Line 81:
 -i 300 --ckptdir /sanscratch/checkpoints/111  ./a.out & -i 300 --ckptdir /sanscratch/checkpoints/111  ./a.out &
 [1] 29201 [1] 29201
 +
 [hmeij@cottontail2 111]$ ps [hmeij@cottontail2 111]$ ps
   PID TTY          TIME CMD   PID TTY          TIME CMD
Line 72: Line 89:
 29210 pts/1    00:00:00 dmtcp_coordinat 29210 pts/1    00:00:00 dmtcp_coordinat
 29212 pts/1    00:00:00 ps 29212 pts/1    00:00:00 ps
 +
 +# terminate half way through
 [hmeij@cottontail2 111]$ sleep 32m; kill -9 29202 29210 [hmeij@cottontail2 111]$ sleep 32m; kill -9 29202 29210
  
Line 77: Line 96:
 cd /sanscratch/checkpoints/111 cd /sanscratch/checkpoints/111
 ./dmtcp_restart_script.sh ./dmtcp_restart_script.sh
 +
 # ps  # ps 
 0 S hmeij    20891 20890  -bash 0 S hmeij    20891 20890  -bash
Line 83: Line 103:
  
 # You must make sure the old directory and file exists, otherwise # You must make sure the old directory and file exists, otherwise
-[40000] ERROR at fileconnection.cpp:737 in refill; REASON='JASSERT(jalib::Filesystem::FileExists(_path)) failed'+[40000] ERROR at fileconnection.cpp:737 in refill;  
 +REASON='JASSERT(jalib::Filesystem::FileExists(_path)) failed'
      _path = /sanscratch/111/fid.txt      _path = /sanscratch/111/fid.txt
 Message: File not found. Message: File not found.
Line 90: Line 111:
  
  
-# launch new coordinator on random port, log port, 24 hour checkpoints  +# The process will pick up from last checkpoint 
-# make sure you create destination dir for checkpoints+# and write output to original work directory
  
-dmtcp_launch --new-coordinator \ +</code>
-  --coord-port 0 --port-file port.txt --interval 86400 \ +
-  --ckptdir /sanscratch/checkpoints/111 \ +
-  time ./a.out+
  
- 
-</code> 
 ==== Quick-Start Guide ==== ==== Quick-Start Guide ====
  
cluster/190.txt · Last modified: 2020/09/28 07:38 by hmeij07

Page Tools

Except where otherwise noted, content on this wiki is licensed under the following license: CC Attribution-Share Alike 4.0 International
CC Attribution-Share Alike 4.0 International Donate Powered by PHP Valid HTML5 Valid CSS Driven by DokuWiki