RT_SCTEMP/job_INSERT_INVOICE_SECTION_STATUS_CABS.100556.fifo

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
pilli
Participant
Posts: 7
Joined: Wed Mar 19, 2008 5:58 am

RT_SCTEMP/job_INSERT_INVOICE_SECTION_STATUS_CABS.100556.fifo

Post by pilli »

Im running a sequencer where a single job is called in mulitple invocations. The job name is job_INSERT_INVOICE_SECTION_STATUS_CABS. For around 10 invocations the following error is encountered.

RT_SCTEMP/job_INSERT_INVOICE_SECTION_STATUS_CABS.100556.fifo
LOCKED STATUS () -1); file is locked

Please advice.

Thanks.
ray.wurlod
Participant
Posts: 54607
Joined: Wed Oct 23, 2002 10:52 pm
Location: Sydney, Australia
Contact:

Post by ray.wurlod »

It would appear that two jobs have generated the same pipe filename. This is odd, as the numeric component should be the process ID. But, if you are using an O/S with large pids and the name only uses the rightmost six characters, I guess there's a chance of this happening. Involve your support provider; if my analysis is correct you've uncovered a bug.
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
kumar_s
Charter Member
Charter Member
Posts: 5245
Joined: Thu Jun 16, 2005 11:00 pm

Post by kumar_s »

What is the invocation Id for the Current run of this job?
Impossible doesn't mean 'it is not possible' actually means... 'NOBODY HAS DONE IT SO FAR'
verify
Premium Member
Premium Member
Posts: 99
Joined: Sun Mar 30, 2008 8:35 am

Post by verify »

Hi Ray/Kumar,

Thanks for your response. The invocation id for the current run of job is 100556. Does this imply that I am using the same invocation id for 2 jobs due to which I am getting this error. Please confirm my understanding.

Thanks.
RK Raju
ray.wurlod
Participant
Posts: 54607
Joined: Wed Oct 23, 2002 10:52 pm
Location: Sydney, Australia
Contact:

Post by ray.wurlod »

It would appear that such may be the case.
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
kumar_s
Charter Member
Charter Member
Posts: 5245
Joined: Thu Jun 16, 2005 11:00 pm

Post by kumar_s »

Or also may be not. Because, today Iam facing similar issue with non multiple instance job. This job was hung for a long time. Asked Unix people to release the lock. Not sure how they did that. Bit strange. :?
Impossible doesn't mean 'it is not possible' actually means... 'NOBODY HAS DONE IT SO FAR'
ray.wurlod
Participant
Posts: 54607
Joined: Wed Oct 23, 2002 10:52 pm
Location: Sydney, Australia
Contact:

Post by ray.wurlod »

Would you mind asking them?
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
kumar_s
Charter Member
Charter Member
Posts: 5245
Joined: Thu Jun 16, 2005 11:00 pm

Post by kumar_s »

Yeah, first task for tomorrow. The job is locked as well. I cant even delete and reimport.
Impossible doesn't mean 'it is not possible' actually means... 'NOBODY HAS DONE IT SO FAR'
verify
Premium Member
Premium Member
Posts: 99
Joined: Sun Mar 30, 2008 8:35 am

Post by verify »

kumar_s wrote:Yeah, first task for tomorrow. The job is locked as well. I cant even delete and reimport.
Dear Ray/Kumar,

I am running around 100 jobs in parallel of which 50 jobs are getting locked due to the above fifo file locked error. What could now be the possible solution for the above issue.

Thanks.
RK Raju
ray.wurlod
Participant
Posts: 54607
Joined: Wed Oct 23, 2002 10:52 pm
Location: Sydney, Australia
Contact:

Post by ray.wurlod »

Still awaiting response from kumar_s...

Is yours precisely the same error, with precisely the same number in the name of the file?

If not, are the numbers in your file names corresponding to process IDs or, perhaps, invocation IDs of multi-instance jobs?
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
pilli
Participant
Posts: 7
Joined: Wed Mar 19, 2008 5:58 am

Post by pilli »

ray.wurlod wrote:Still awaiting response from kumar_s...

Is yours precisely the same error, with precisely the same number in the name of the file?

If not, are the numbers in your file names corresponding ...
There are around 200 jobs which are being called with invocations. 80 jobs are being aborted with the fifo file locked error.
eg: RT_SCTEMP/JobName.100556.fifo
LOCKED STATUS () -1); file is locked

The number in the file name is invocation id.eg 100556. I have checked code and concluded that there are no jobs which are being called with the same invocation id.
Please advice on how to solve this issue.

Thanks.
verify
Premium Member
Premium Member
Posts: 99
Joined: Sun Mar 30, 2008 8:35 am

Still stuck with fifo

Post by verify »

Hi Ray,
I searchd thru the posts nd found some advce regarding rights on the RT_SCTEMP dir. As suggested in this posts, I gave all the read,write,execute rights to this directory. In fact, I emptied this dir (by takin a backup) & ran the sequencer again. Bt the error is still coming for a new set of jobs. I also noticed that all those fifo files which are present in this RT_SCTEMP dir, corresponding jobs to these files are aborting. I am totally clueless abt this. Plz help, thr is a large no. of jobs failing due to this err.
RK Raju
chulett
Charter Member
Charter Member
Posts: 43085
Joined: Tue Nov 12, 2002 4:34 pm
Location: Denver, CO

Post by chulett »

Time to involve your official support provider, it seems.
-craig

"You can never have too many knives" -- Logan Nine Fingers
kumar_s
Charter Member
Charter Member
Posts: 5245
Joined: Thu Jun 16, 2005 11:00 pm

Post by kumar_s »

Issue in my site got sorted out. And it turned out to be a dump issue.
The first time when I asked to release the lock for the jobs, they have killed the PIDs from Unix. I think that could left the lock of that job over the RT_SCTEMP folder.
Yes that fifo were present, if the job gets aborted.
Next time, they used Admin window to release the lock of that job. And not its running fine.
I wanted you to check, if you run the job in less number of instance, does it get aborted? Start with 10s and gradually increase with multiples of 10.
Impossible doesn't mean 'it is not possible' actually means... 'NOBODY HAS DONE IT SO FAR'
Post Reply