This is a video replay of our Tuesday, May 10 @ 11 AM ET CA Workload Automation Community TechTalk.
Watch this video to better understand which agent log files to leverage, and what to look for in the various agent log files in order to minimize the effects of workload failures.
Uncover the why, where and how of the CA Workload Automation Agent logging functions.
Speaker: rasgr01 , Sr. Principal Consultant, CA Technologies
I tried to locate the presentation used in this video but couldn't. Could someone provide me the link/location for the presentation. Thank you.
We had this same issue I opened a ticket and a DAR (request for enhancement) but ended up creating my own solution.
Describe the Enhancement Request in detail.
For each job run through ESP a directory is created (APPLICATION.GENERATION) under spool directory in the ESP Agent Directory.
On the filesystem type we have on Linux there is a limit of 31,998 files that can be created in a directory.
We would like to configure the ESP Agent to keep at a minimum of 10 days of files. When the daily volume averages over 3,100 we run into an issue of not being able to write the spool file, since the file system limit has been met. This causes the agent to stop working.
At every interval where the ESP Agent checks for spool files that meet the expiration time and deletes them, it should also check directories and deletes them if they are empty.
Additionally it would be helpful if when starting the ESP Agent would check the spool directory file system to determine how many directories can be created. At every interval when it checks for spool file cleanup it should determine what percentage of capacity is used and it should send an alert when that number exceeds a threshold.
Do you have an available workaround? If yes, what is it?
I have set agent.spool.success.autocleanup=true and created a shell script that checks for empty directories and deletes them. Currently job is scheduled to run daily.
Another solution would be to change the filesystem type which would increase the limit to the number of files created. On the impacted servers the filesystem type is ext3, if you change to ext4 it increases the 32,000 subdirectory limit in ext3 to unlimited.
Let me know if you are interested in seeing the ESP PROC and shell script that I created.
I have set agent parameters "oscomponent.joblog.success.autocleanup=true", its not clearing the spool directory.