Job Control fatal error (-14)
Moderators: chulett, rschirm, roy
Job Control fatal error (-14)
I have a problem whereby the job was aborted. I checked the Log and it indicate the following error message:
Batch::<Batchname>.JobControl(fatal error from DSRunJob): Job control fatal error (-14)
(DSRunJob) Job <jobname> apprears not to have started after 60 seconds.
Kindly advice.
Batch::<Batchname>.JobControl(fatal error from DSRunJob): Job control fatal error (-14)
(DSRunJob) Job <jobname> apprears not to have started after 60 seconds.
Kindly advice.
-
- Participant
- Posts: 54607
- Joined: Wed Oct 23, 2002 10:52 pm
- Location: Sydney, Australia
- Contact:
It's not always about how Big and Beefy your server is, or the number of jobs running at the same time - it can be about the number of jobs you try to start at the same time.
Your 'abnormal termination' errors are a separate issue. Suggest you start a new thread and include as much detail as you can on it, including what you get when you Reset the job and find a 'From previous run...' entry in the log.
Your 'abnormal termination' errors are a separate issue. Suggest you start a new thread and include as much detail as you can on it, including what you get when you Reset the job and find a 'From previous run...' entry in the log.
-craig
"You can never have too many knives" -- Logan Nine Fingers
"You can never have too many knives" -- Logan Nine Fingers
-
- Participant
- Posts: 94
- Joined: Wed May 08, 2002 8:44 am
- Location: Germany
- Contact:
As Ray mentioned this is usually due to the fact that you are overloading your system. Means you are starting too many processes at the same time. If a process could not be scheduled 60s after being initiated, DS terminates it.
You should try to more serialize your sequence flow.
You may also try to ask IBM (Ascential) support if they can provide you a patch that extends the timeout period. I know this has been provided for some UNIX systems, not sure for Windows....
Best regards
Klaus
You should try to more serialize your sequence flow.
You may also try to ask IBM (Ascential) support if they can provide you a patch that extends the timeout period. I know this has been provided for some UNIX systems, not sure for Windows....
Best regards
Klaus
Hello again Klaus,
are you certain about that patch?
I am at a site where they had this problem and I was told here that there was no workaround for this issue; there were some configuration changes done to allow more processes in total but that the 60 second limit was hardcoded in several places and couldn't be changed. This is an AIX site. It would be great if this truly did exist.
are you certain about that patch?
I am at a site where they had this problem and I was told here that there was no workaround for this issue; there were some configuration changes done to allow more processes in total but that the 60 second limit was hardcoded in several places and couldn't be changed. This is an AIX site. It would be great if this truly did exist.
<a href=http://www.worldcommunitygrid.org/team/ ... TZ9H4CGVP1 target="WCGWin">
</a>
</a>
-
- Participant
- Posts: 54607
- Joined: Wed Oct 23, 2002 10:52 pm
- Location: Sydney, Australia
- Contact:
Additional Suggestions: Disable any screen saver on the Windows server. Verify that the firewall and/or anti-virus software is not imposing too great an impact on startup times for executables. Verify that the Windows server is not being used as either a primary or backup domain controller. Monitor resource consumption with Task Manager or Performance Monitor to "spot the hogs".
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
-
- Participant
- Posts: 94
- Joined: Wed May 08, 2002 8:44 am
- Location: Germany
- Contact:
Hello Arnd, hope you're doing well
Yes, I've got that patch for a SOLARIS and an AIX customer, both on release 7.51A.
I think they don't want to provide that patch in general, but on insisting specific customer request it was provided and helped resolve their issue.
Greetings from good old Bavaria...
Klaus
Yes, I've got that patch for a SOLARIS and an AIX customer, both on release 7.51A.
I think they don't want to provide that patch in general, but on insisting specific customer request it was provided and helped resolve their issue.
Greetings from good old Bavaria...
Klaus
Klaus,
thanks for that information & gruesse right back! I'll pass on the information here to get the customer to request that patch; they are suffering from that issue and their workarounds are not permanent solutions.
Danke,
thanks for that information & gruesse right back! I'll pass on the information here to get the customer to request that patch; they are suffering from that issue and their workarounds are not permanent solutions.
Danke,
<a href=http://www.worldcommunitygrid.org/team/ ... TZ9H4CGVP1 target="WCGWin">
</a>
</a>
Thanks for the pointers Klaus. We have jobs running in parallel. Around 6-7 jobs at a time. That might be one of the problem, as you mentioned earlier. Just curios, is there a limit whereby Datastage could handle an "X" amount of parallel jobs running or is it based on the 60 seconds window time frame?
-
- Participant
- Posts: 54607
- Joined: Wed Oct 23, 2002 10:52 pm
- Location: Sydney, Australia
- Contact:
How long is a piece of string? The answer is "it depends". More jobs can be handled at once if they don't do much compared to fewer jobs that each do a lot. If you make extensive use of common hashed files, you might investigate sharing these (see dsdskche.pdf in your manuals). Ultimately try to design jobs so that they do the minimum amount of processing and I/O to achieve the desired result.
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
Iam not at PX right no, but I am sure, there is a Evnironmental variable, where you can try setting it for time outs.
Last edited by kumar_s on Mon Jun 26, 2006 12:48 am, edited 1 time in total.
Impossible doesn't mean 'it is not possible' actually means... 'NOBODY HAS DONE IT SO FAR'