No successful initialization of the delta took place

[Former Member]

This morning I am suddenly getting this error in many of my process chains;

"No successful initialization of the delta took place"

I'm baffled as I can't think of anything I changed in the system since the last successful day which was Friday morning. This message is not very useful; all of the 10+ infoproviders that have failed (with same message) have been working for a long time and have indeed had successful delta initialization requests so this makes no sense.

Anybody experience this or now what I should look into to fix this? I tried rerunning immediately in foreground and I get the same error.

[Former Member · Accepted solution]

Hi patrick,

The init flag for all those IP's must have been deleted. Please check the same for the IP in scheduler->initialization option for source system. I dont think there is any entry here and somebody by mistake have deleted this.

Sriram

[Former Member · Accepted solution]

But I can see the INIT request package when I go to MANAGE any of the problematic cubes. It seems they are there but the update doesn't recognize them for some reason. We don't have anybody here who would have deleted them.... although we have had a consultant dial-in to diagnos some poorly running queries.

[edwin_harpino · Accepted solution]

hi Patrick,

since it's happened to 10+ infoproviders, i think there may PI support package upgrade in r/3 side ? cross check with basis guy ? in r/3 rsa7 or smq1 can you see these datasources queue ?

hope this helps.

[Former Member · Accepted solution]

When I go to the info package and click scheduler/initialization options you are correct I do not see an initialization request but I can see each cube has the request sitting in the cube as the first request. How could somebody have deleted all of those at once? That seems unlikely. I'm the only developer here at my company besides the consultant that dials in to help.

[Former Member · Accepted solution]

Patrick

As mentioned check any activities happend over the week end in R/3 side like applying support Packs or any other development.

Thnaks

sat

[Former Member · Accepted solution]

Patrick,

look at SLG1 to verify from the log if your consultant...

(if you want to know the exact object id, do a little test on your development system: create a new init and then delete it and verify what is the generated log...)

Are you performed some basis activity on your source system?

Hope it helps!

Bye,

Roberto

[Former Member · Accepted solution]

Ok we looked and there are no delta queues appearing! I don't understand how that could have happened.

[Former Member · Accepted solution]

Thank you Roberto, the log definitely indicates somebody deleted the init requests!!! It shows a consultant name though who hasn't been here for ages. My boss says that many jobs will still appear under this consultants name because he is the one who last saved the objects or something like that. So I'm not clear on who or how it happened. Is there a way to fake the system into thinking the inits are there now by performing a 'simulated init request'?

[Former Member · Accepted solution]

Hi again

you can run a init without data transfer to all ur targets so that there will be a init flag. But what happens here is u lose the data which is updated in between. The delta start coming from the point where u carried this new init. So what i suggest is go for re-inits.

Sriram

[Former Member · Accepted solution]

First of all, check if this old user runs some (scheduled) job...

Anyway, the only (not so easy) solution is to do some repair request (since you loose all delta created during this "black hole")...

Otherwise, I think you have a lot of work to do...

Bye,

Roberto

(and please don't forget to reward the answers...it's THE way to say thanks here !)

[Former Member]

Patrick

Please check Initialization in your info package before you do further analysis may be by mistake some one might have deleted.

Thnaks

sat