DSXchange

One of the sequencer( underneath 5jobs) got triggered for every 5 mintues. Genreally it takes max of 1-2 min for completion of execution. But sometime any one of the job got hung and it runs continuously.
If i clear the resources from director, job get aborted and again in the next of couple of runs it gets hung.
After abortion of job the processes unix as below.
Can anyone suggest to get rid of this? I can see phantom processes and also .fifo file which might have corrupted. What can be done to resolute this?[/code]

What you really need to do is to address the reasons why the underlying jobs exhibit this intermittent behaviour.

A job sequence can not complete until all of the activities in it have completed (or a Terminator activity is executed), therefore if the underlying activity is hanging, or simply taking a long time to execute, then the job sequence itself will patiently await its completion.

Killing underlying processes is fraught with a number of dangers, primarily that you do not identify all the processes associated with the job - for example external commands (whether from the job sequence or from stages such as External Filter) do not execute osh and therefore are not picked up by the Cleanup Resources tool (which was - and really still is - intended primarily for server jobs).

If you are unable or unwilling to determine and prevent the problems in the underlying activities, then I would suggest that you create custom job control code that establishes a time within which the activity must complete and issues a stop request if this timeout occurs. You can use the generated job control routine from the job sequence as a template and paste it into the Job Control tab of a new server job as a prototype that you can edit.

solution -

Once we set these 2 paramters in user defined section of Administrator, the issue is resolved.

APT_DONT_OPTIMIZE_MODIFY=1
APT_MSG_LINE=1

Not sure actual reason. If anyone have any clue on this...please share with me.
But it is always suggestable to clear the &PH& directory periodically for a project to get rid of the issue.

Thanks

Please explain what these environment variables do, and how you came to decide on these specific settings for them, so that we all might understand your resolution and why it may have worked.

APT_DONT_OPTIMIZE_MODIFY=1
APT_MSG_LINE=1

I applied these two values then also the &PH& files are getting created any idea, how to solve this so that it doesnt create any &PH& files.

No can do, those files are essential to the proper operation of the product.

Could you please tell me how it is important. bcos we are planning to purge this on a timely basis.

Very. Purging or pruning them is fine, however, as long as the job they are associated with has completed running. Typically, people script something to run every day and remove any items X days old where X is on the order of 3 to 7, I would guess.

Can we do something like remove all the files in &PH& directory?
is this a show stopper?

Only if you are certain that no jobs are running at that time.

Why so worried about this?

yes thanks,

Can you send me any document or anything if you have, to know what &PH& is all about, bcos yesterday it had taken 2GB of data.
So being little proactive here. Thanks

Yah, sorry... I should have noticed the hijacking in progress but we're halfway to Cuba by now.

Hi Ray,

I am new to this forum, i dont know exactly how to post it, so i open one top and click post new topic and continue, will check and do it correctly.

Top left of page, "Post new topic". Make sure you are in the appropriate forum first.