COGIPF_AGENTRUN filled with repeated messages

[Sunchaser]

Dear all,

Here is a "small" problem I'm not able to solve, and in addition Cognos now says that there is no support for C8.4 anymore.

After a problem with a report "blocked", due to the size of the PDF generated that seemed to exceed the capacity of the SMTP server (even if the local IT support contacted  for the case says actually that it is not possible), I had to kill the hanging process.
But I saw that in the table COGIPF_AGENTRUN (we have DB Cognos log set to "basic", etc ..), error messages were still recorded; here it the error:
"
CNC-SDS-0335 Error returned from SMTP Server. The error message was: [552 5.3.4 Message size exceeds fixed maximum message size ]. The message could not be sent.
"
Even after a restart of the Cognos service, the messages are still recorded every 30 minutes (approx), even when nothing is running.
The amount of records is growing day after day, it is unnecessary and I don't like that.

Do you know where it is coming from ?

Thanks.
Best regards
V.

[Grim]

Someone is attempting to email a report that is too big for your mail server.

[Sunchaser]

Someone is attempting to email a report that is too big for your mail server.

Ok ... sorry to be a little bit negative, but - obviously - "of course" I understood this.
I know & I have identified two weeks ago that it was coming from an "error" of a user who produced a PDF so big that it was not possible to be sent per mail.
Ok.
The hanging process in Cognos was killed.
Then, as the messages were still there and recorded regularly, we restarted Cognos.
But, when Cognos started again, the phenomenon reappeared.
So, even when Cognos is not used (no interactive process, no background process), the COGIPF_AGENTRUN table is still receiving records every 30 minutes approx.
Funny, isn't it ?
I don't have the hand on the SMTP server, I have to ask whatever could be necessary about this to an "IT service". But they said that it was ok from their side.
So I wonder where do this bloody records are coming from ?
As soon as Cognos is up and running, this records are inserted again, again, again ...

So, where could this come from ?

Thanks,
Regards
V.

[Grim]

I should have been a little more detailed. My apologies. I focused on the error message and skimmed the rest. My bad. So with a bit more detail this time...

When the messages are being put into the AGENTRUN table I believe that this means that there is a "job" that is scheduled to "run and send" the report. So you may want to try to identify the job, and not just the report. Then cancle/delete/disable that "job" and see if that helps clear up the errors.

[Sunchaser]

Dear Grim,

Thank you for your answer.
I have checked again, and ... no, there is nothing scheduled. Not "officially" at least.
I checked in Cognos, nothing of this kind. The schedules existing are not related to the report flagged in the error message, and they can be stopped without any effect on the messages in AGENTRUN.
Nothing of this kind in an other application.
No job at database level, and there is no other third party application that could ask for a report.

My assumption would be that there is something recorded in one of the tables of the content store.
Something that we don't have access to through the "normal" interface (IBM Cognos Administration), because it has been "deleted", but due to any kind of problem or bug, one record is still existing at database level and it is pushing Cognos to try to resend the report endlessly.
This is only a assumption, not sure it is correct / plausible of course.
I'll see soon, as we will restart the productive server this week end.

Best regards
V.
 

[Sunchaser]

Hi,

Even after restart of the server, even if all schedules stopped in Cognos, even if no job is running at database level (or anything of this kind), the messages are still recorded, every 30 or 40 minutes ...
Ghost in the system ?