BIA Error 2625 & 2613

[Former Member]

Hi Experts,

We are using BIA rev. 47 and BI SPS12. Sometimes when we fill the index for an InfoProvider, and at the same time run a query on a MultiProvider based on the mentioned InfoProvider, we get some error messages, and no data is fetched.

The error messages are:

Error while reading data, navigation is possible.

Errors occurred during parallel processing of query

Error reading the data of InfoProvider <name of BIA index>

The following error 2625 occurred in BIA server

.....

The following error 2613 occurred in BIA server

Error executing physical plan: executor: communication problem

Sometimes the infoproviders mentioned in the error message are not even the same as the one which index is being filled, sometimes the InfoProviders reffered to are the other InfoProviders used in this query.

Have anybody else seen these errors, and do you have any solutions on how to solve them or further investigate them? They do not create any dumps in the system which could have been analyzed further.

Thank you in advance,

Mikael

[former_member93896 · Accepted solution]

Hello Mikael,

errors 2613 and 2625 are side effects of TREXIndexserver not being available. Please open a customer message so SAP support can analyze why the index server crashed.

Regards,

Marc

SAP NetWeaver RIG

[Former Member · Accepted solution]

Hi,

Thank you for your suggestions. We have opened an OSS message to SAP in order to hear their opinion.

Best regards,

Mikael

[jgleichmann]

Hello Mikael,

update to Rev48, here the services are more stable than in Rev47. Have a look at Note 1231097 ( BIA 7.00: Revision 48 ).

Best Regards,

Jens

[Vitaliy-R]

Jens,

When was this Revision released? Yesterday (note is updated on Oct 15)? That's a good news.

Thanks,

-Vitaliy

[jgleichmann]

Hi Vitaliy,

Rev48 was released on Oct 15 in the evening. I installed it yesterday morning(Oct 16) and till now it seems to be that many of my problems I had in Rev47 are solved.

Problems I had:

- After create new indices there will be an initial reorg and if the process chain use this index, it will be blogged and the process chain abort.

- no stable services, many aborts from the nameserver services

- I can´t run with backup mode mutual, the take-over from the backup note worked not fine

Till now the behaviour is good, but after 1 day this is not representative

Best Regards,

Jens

Edited by: Jens Gleichmann on Oct 17, 2008 8:23 AM

[former_member184494]

Mikael,

With regard to unrelated cubes - we usually found that is some other cube using the same characteristics is being used then this issue used to happen - but later on when re executed , the issue went away...

Hope it helps...

Arun