AutoSys Workload Automation

A specific job delays its 1:00 a.m. start for 3 to 5 minutes because of "startjobfail" alarm, then runs fine.  Other jobs appear to run fine at 1:00 a.m. without this issue.  Need help troubleshooting what could be causing this issue with this particular job.  Currently running Autosys 11.3.6. inc 1 on the servers.

Anyone have any thoughts as to why this is occurring for this specific job or where to start?

Hi,

What exactly is the specific job - can you provide the job definition?

Does this job run multiple times a day and only during the 1am job run, the job is delayed?

Within the event_demon log, are there any other CAUAJM_E messages in regards to that specific job? Anything related to the machine it is supposed to execute on or a starting condition not being met?

Thanks, Jameela

I have opened a ticket with CA and we are working the issue.  At first was thinking it was a security issue on a specific server because of an error in the logs but after resetting the Owner account password on that specific server, it did not help.

Patrick, were you able to work through the issue with support?

Is this a windows job? is this on a heavily used machine?

I have seen such issue occur it if its a heavily used machine it could be an issue with the number of open files/processes.

if this is a dbproc jobtype it could be a setting you need to update in agentparm.txt It has to do with the threads etc, need to set to 0.

if we can get more specifics those of us that have lived through such issues may help.

Hope this helps and good luck

Steve C.

Have found the cause of the issue.  Worked with CA Tech and a member of our Windows Server support team.

We recently updated our settings for what servers that the service accounts (owner) could log on to.  Previous audit finding that had to be corrected because some of our Autosys Owner accounts could access servers that were not required to process jobs.  Someone mistyped the server name on this particular Owner account for one of our application servers. So what was happening is that the owner account was trying to access that specific application server and could not.  After so many tries, it would go to another application server and process the job to completion.

thanks.