Error setting up internal communications

Post questions here relative to DataStage Enterprise/PX Edition for such areas as Parallel job design, Parallel datasets, BuildOps, Wrappers, etc.

Moderators: chulett, rschirm, roy

Post Reply
anitha_casturi
Participant
Posts: 21
Joined: Wed Jun 30, 2004 3:48 am

Error setting up internal communications

Post by anitha_casturi »

Hi,

We have been facing this issue in the QA environment.

Error setting up internal communications (fifo RT_SCTEMP/job.fifo)LOCKED STATUS () -1); file is locked

The job does not seem to be locked but while we execute this job, we run into this issue. Couple of days back we faced this issue for another job and we got the administrator to unlock.

Is there anyway we can prevent this issue?
Thanks,
Anitha
ray.wurlod
Participant
Posts: 54607
Joined: Wed Oct 23, 2002 10:52 pm
Location: Sydney, Australia
Contact:

Post by ray.wurlod »

The fifo files are named pipes used (probably) for inter-process communication. Perhaps there are left-over fifo files from an earlier aborted job? Could it be that two concurrent jobs are trying to use the same fifo file?
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
anitha_casturi
Participant
Posts: 21
Joined: Wed Jun 30, 2004 3:48 am

Post by anitha_casturi »

Ray - Thanks for the response.

We have all the jobs running in multiple instances but the various instances are identified using Unique invocation id.

The job that has got locked, did not abort the last time.

Alternately, we tried to delete the fifo file and tried trigerring the job again, but still we are running into the same issue
Thanks,
Anitha
ArndW
Participant
Posts: 16318
Joined: Tue Nov 16, 2004 9:08 am
Location: Germany
Contact:

Post by ArndW »

Have you made sure that there are no left-over processes which might still be attached to the FIFO file. The easiest guaranteed solution is to bounce the server, if that is possible. Otherwise check to see if you have any processes left running when nothing is happening on the machine. Alternately, save the job under another name, delete the original, then rename the copy and compile it. This will give the job a new job number and then you can see if the error persits when you re-run the job.
ray.wurlod
Participant
Posts: 54607
Joined: Wed Oct 23, 2002 10:52 pm
Location: Sydney, Australia
Contact:

Post by ray.wurlod »

Does the name of the fifo file include just the job name, or does it include the invocation ID as well? If it's only the job name, then you've bumped up against a limitation of multi-instance running.
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
anitha_casturi
Participant
Posts: 21
Joined: Wed Jun 30, 2004 3:48 am

Post by anitha_casturi »

ArndW-Thanks for the response. Well bouncing the server may not always be possible in a production environment. Is there any way, we can avoid encountering this error?


Ray-The FIFO file has the name fo the job and the invocation ID as well.
Thanks,
Anitha
ArndW
Participant
Posts: 16318
Joined: Tue Nov 16, 2004 9:08 am
Location: Germany
Contact:

Post by ArndW »

Since we don't know the cause of the problem we can't give a solution to ensure that it doesn't happen again.
Post Reply