jobs going to held status though there are free slots availa

[ulab]

HI Experts,

I see few jobs go into held status though there are plenty of free slots available in datastage grid, the wonder is that jobs with xxxxxx NUID which is executing fine for some time is going into held status even though there are plenty of free Slots available, this happens only for few jobs sometimes in a day, kindly please let me know if you have any answer on this,


Please let me know if you need more information,

thank you,
Ulab

[ulab]

the below is some more info on my post..hope this helps more to understand,

Prompt> llclass
Name MaxJobCPU MaxProcCPU Free Max Description
d+hh:mm:ss d+hh:mm:ss Slots Slots
--------------- -------------- -------------- ----- ----- ---------------------
dsbatch undefined undefined 36 62 Class for DataStage batch jobs

Id Owner Submitted ST PRI Class Running on
------------------------ ---------- ----------- -- --- ------------ -----------
dseax2121.283205.0 oprddw 8/6 03:45 R 50 dsbatch etlax2103
dseax2121.283334.0 oprecf 8/6 04:15 R 50 dsbatch etlax2105
dseax2121.283336.0 oprwfm 8/6 04:15 R 50 dsbatch etlax2103
dseax2121.283339.0 oprwfm 8/6 04:15 R 50 dsbatch etlax2104
dseax2121.283341.0 oprpsa 8/6 04:16 R 50 dsbatch etlax2104
dseax2121.283365.0 oprpsa 8/6 04:19 R 50 dsbatch etlax2104
dseax2121.282533.0 oprfss 8/6 01:57 H 50 dsbatch
dseax2121.282735.0 oprpsa 8/6 02:23 H 50 dsbatch

[ulab]

any suggestion please.

[chulett]

Contact your official support provider. Here you can only be helped by people with grid experience and I don't think that's a very high number.

[ray.wurlod]

Is the grid manager process running?

[ulab]

Thank you for your responce Ray and chullet,

the grid is running perfectly fine and all the resource nodes are active, the only surprise for me is the same USER is able to run a job at one point and the same user did not able to get a reqource node though there are 70% of free slots are available,

I was breading my head, as a alternate just am canceling the job which were in queue, and asking the application team to run the job again,