Differences

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

--- doku:epilog [2014/06/03 12:44] – [How to use] ir
+++ doku:epilog [2021/05/13 17:45] (current) – removed goldenberg
@@ Line 1: / Line 1: @@
-====== User-specified epilog scripts ======
-===== Synopsis: =====
-After job termination, epilog scripts are executed on all compute nodes where the job was running on.
-This enables **copying back node-local temporary data (/work/tmp)** whether or not the job is killed by a runtime limit.
-===== How to use =====
-A bash script called **"epilog.sh"** has to be placed into the same directory you submit your job from (working directory, $SGE_O_WORKDIR). After job termination, this script is executed **on all of the job's nodes**. A user defined script name can be used within the job file (it's probably best to specify the full path):
-<code>#$ -v USER_EPILOG=/path/to/my_epilog_script.sh</code>
-or when submitting the job:
-<code>qsub -v USER_EPILOG=/path/to/my_epilog_script.sh job</code>
-**Note***:
-It is absolutely required that **only the minimum necessary work is done in an epilog script**.
-Long-running tasks in an epilog file will not be tolerated! There is a timeout to prevent buggy epilog scripts from running forever.
-==== Example epilog.sh script ====
-<code># point this to where your program writes its data to. you should include $JOB_ID, $SGE_TASK_ID, $QUEUE and $HOSTNAME to make sure the directory is unique.
-TMP_DIR="/work/tmp/${QUEUE}_${JOB_ID}_${SGE_TASK_ID}_${HOSTNAME}"
-cp -r $TMP_DIR ${SGE_O_WORKDIR}</code>