on 04-22-2009 9:14 PM
I have found a lot of threads on this already, but none that really answers my problem. I have an attribute change run that fails occasionally. It's not consistent - maybe just like every 10 days. When I go into the program RSDDS_CHANGERUN_MONITOR to reset the run it says "4 aggregates need to be edited." Does anyone know what this means? The attribute run is in a pc for the SAP supplied technical content process chains. And these are the supplied aggregates, so I don't understand what the issue is.
So, my question more has to do with how I can adjust these aggregates so that the changerun doesn't fail? It's a hassle to have to delete the change run, then deactivate the aggregates, then re-run with no aggregates, then reload the aggregates after every failure and then re-run the pc one more time after every failure.
We are on BI 7.0 SP 17
Thanks in advance,
Brett
Edited by: Brett Hutchinson on Apr 22, 2009 10:14 PM
the percentage change of master data can be set in TCODE RSCUSTV8. I see this in a BW3.x system have not tried it though - worth a shout...
I think this sets the change run option for an aggregate based on the percentage change in masterdata based on which you want to trigger a change run to an aggregate...
Edited by: Arun Varadarajan on Apr 23, 2009 2:07 AM
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Thanks for the quick response. The wait time parameter may help us. I will have to test first though. I thought someone somewhere said that parallel change runs in 7.0 were no longer an issue though. Do you know if that is true? Also, do you know what kind of entry I should put in there. Example: If I enter 5, does that mean 5 minutes or 5 hours or 5 seconds? It doesn't seem to say in the help section.
Page 26
Edited by: Arun Varadarajan on Apr 23, 2009 2:25 AM
Have a look at the OSS Note: 825927. Whether this could help you.
Sunil
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
When I go into the program RSDDS_CHANGERUN_MONITOR to reset the run it says "4 aggregates need to be edited." Does anyone know what this means?this means that 4 aggregates have to be undergone change run.
Normaly change run fails if we have another running in the system.Check the timing of the change run,Is their any back up activites are going on at that point of time.
Some time user loacked is alos the issue for the same.
RSCUSTV8 setting is used if your change run is taking long time pls change this setting as below
limit with delta 20
and the block size as maximu as 999999999
this will greatly help you!
Thanks
NS
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
A reason for this may be your infoproviders which are included in the attribute change run step are getting filled by another process chain or manually by some one . Please find out when the att change run step fails whther a load is going on in the infoprovider included in your att change run step .
I also had faced a similar problem and this was the reason for failure.
Thanks for the response. I have scheduled this particular process chain at a time when nothing else runs to make sure that there are no other change runs taking place during the chain. I also implemented cr_maxwait just in case, which didn't help either. This is also the SAP supplied technical content chain (0TCT), so there are no other loads that we do against these info providers. Checking to make sure there are no other change runs happening concurrently was the first thing I checked.
Hello Brett,
Have you the problem solved? We have the same fight for a while, tried everything possible, nothing helped us though. (rebuilt aggregates, set index for the Master Data tables Etc.)
If you had an useful solution, would be thankful.
Dóra
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Sorry, I still haven't found anything either. It's completely random here, but I can get the pc to work if I deactivate the aggregates, clear out the failed change run (in sa38/se38 by running RSDDS_CHANGERUN_MONITOR), reactivating the aggregates and then re-running the pc. It's just a pain having to do it everytime it happens. I will make sure to post something if I figure anything out. I even tried separating the change runs 1 per info object, but that still didn't work. I tried adjusting the wait time in rscustv8 also. I tried using parallel proessing in the change run variant also. Nothing has worked. Please let me know if you find anything also.
| User | Count |
|---|---|
| 3 | |
| 3 | |
| 2 | |
| 1 | |
| 1 | |
| 1 | |
| 1 | |
| 1 | |
| 1 | |
| 1 |
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.