cancel

Schedule Event Triggers seem to have stalled

Highlighted
DyN
Regular Contributor.

Schedule Event Triggers seem to have stalled

I am running HP TRIM 7.10.1132.  Made a lot of changes to retention schedule triggers, and it seems to have ground my services to a halt.  I have tried relocating the queued events to a less used workgroup server (I have three available to me) but hasn't made a difference.  It seems like it is processing 1 every other day or so.  I have also tried stopping and restarting the services to no avail.  Am I stuck waiting this out, or is there a way to trouble shoot what is going on using Enterprise Studio?

8 REPLIES
Grundy
Acclaimed Contributor.

Re: Schedule Event Triggers seem to have stalled

There can be issues processing scheduled event triggers when you have extremely overloaded containers.

We have a recommendation in the TRIM_Spec to keep containers to less then 300 documents in size.

Once you start getting larger, you compound the calculation of schedules against these containers and it brings everything to a grind.

 

Check for these.

Ultimately, you'll need to wait for these slower containers to process or clear the scheduled event trigger events and process them once you are on 7.2.1 patch 3 or later. We've made some improvements to large container processing.



::::::::::::::::::::::
NOT A HP EMPLOYEE
::::::::::::::::::::::

INFORMOTION.com.au
Neil Summers
Acclaimed Contributor.

Re: Schedule Event Triggers seem to have stalled

Try asking your DBA to run the following SQL script to identify containers in TRIM with more than 300 contents that might be contributing to slow schedule event processing:

 

select b.title,

            b.fullrecordid,

            count(a.uri) as #Documents

from TSRECORD a, TSRECORD b

where a.rccontaineruri=b.uri

group by b.title,

            b.fullrecordid         

having (count(*) > 300)

order by 3 desc


Neil

Note: Any posts I make on this forum are my own personal opinion and (unless explicitly stated) do not constitute a formal commitment on behalf of HPE.

(Please state the version of CM you're using in all posts.

HPE Software Support Online (SSO): https://softwaresupport.hpe.com/
DyN
Regular Contributor.

Re: Schedule Event Triggers seem to have stalled

Thanks Neil.  Ran the script with the following results.  Doesn't seem to be the source of the problem.

 

12:01:36 SQL> select b.title, b.fullrecordid, count(a.uri) as "#Documents"
12:01:47   2  from trim.TSRECORD a, trim.TSRECORD b
12:01:47   3  where a.rccontaineruri=b.uri
12:01:47   4  group by b.title,b.fullrecordid
12:01:47   5  having (count(*) > 300)
12:01:47   6  order by 3 desc
12:01:47   7  ;

TITLE                                                                          
--------------------------------------------------------------------------------
FULLRECORDID                                       #Documents                  
-------------------------------------------------- ----------                  
                                                                               
                                                       597111                  
                                                                               
E-Mail Discovery & Retention - PHASE ONE - 6/1/05 to 5/31/06                   
RIM-06/026                                                697                  
                                                                               
FY2007                                                                         
RIM-06/016                                                312                  
                                                                               

Elapsed: 00:00:08.54
12:01:57 SQL> spool off

Grundy
Acclaimed Contributor.

Re: Schedule Event Triggers seem to have stalled

Yeah, only 2 large containers and they aren't that huge.

 

Maybe try splitting of SET processing to a separate server?



::::::::::::::::::::::
NOT A HP EMPLOYEE
::::::::::::::::::::::

INFORMOTION.com.au
EWillsey
Acclaimed Contributor.

Re: Schedule Event Triggers seem to have stalled

What types of changes did you make to the schedule triggers?
DyN
Regular Contributor.

Re: Schedule Event Triggers seem to have stalled

We made many changes to the Rounding of the retention period.  Some were to fiscal year, some to calendar year from an initial setting of no rounding.

EWillsey
Acclaimed Contributor.

Re: Schedule Event Triggers seem to have stalled

When I see this happening at TRIM sites I turn on logging (6R2 just event server, 7.0+ it's at the workgroup server level). If it's on a server not used by anything else that's even better. Inspect the event server log to see what it's processing... and check the workgroup server logs to see the statements it's issuing.

If the problem is with number of items contained, then you'll see lots of workgroup server log entries that look like:

xx:xx:xx 0 xx trimservices GetRecordset(main)[3 2 2 ] SELECT * FROM TSARCHIVEE WITH (NOLOCK) WHERE ( aeRecordUri = 2080083 AND isEventPending = 'T' AND aeEventType = 106)


If you've got a large nested folder structure and you did not setup schedules on the top-levels of the structure, then TRIM could be iterating down the structure a lot for each new item.

Sometimes it's easier to just delete all of the pending SET events and manually recalculate everything.
DyN
Regular Contributor.

Re: Schedule Event Triggers seem to have stalled

Thanks for the advice.  I will look back through the logs to see if there's anything like that going on.  For now, the events have finished processing.  It only took 3 weeks of babysitting the services to complete the 35,000+ events.