function readOnly(count){ }
Sign Up ›
Starting November 20, the site will be set to read-only. On December 4, 2023,
forum discussions will move to the Trailblazer Community.
Learn More

Browse by Topic
ShowAll Questionssorted byDate Posted
+ Start a Discussion
IntegrationsIntegrations 

Informatica jobs suddenly failing

We have been running in a production environment for over a month.  Suddenly last night all of our informatica jobs started failing with the following error.
 
ERROR: Session task instance (s_m_Name_of_Mapping): Execution terminated unexpectedly.
 
That's it.  Nothing more.
 
We were running jobs yesterday and everything was working fine until last nights load.
 
Any idea's??
SuperfellSuperfell
You need to get a patch from informatica, they made some incorrect assumptions about how long a sessionId could be, and the winter 08 release changes that.
IntegrationsIntegrations

I shared this with my resident Informatica administrator and here were his comments...

I don't believe this is the issue we are having. The sessionID that Simon talks about is when you have connected to Salesforce and then the ETL goes off and does something else and when it tries to come back to Salesforce it has effectively timed out.

The issue we are seeing is that we connect but the first read of data from Salesforce, we get the message "Execution terminated unexpectedly."

Any other idea's????

Message Edited by Integrations on 11-16-200710:01 AM

Message Edited by Integrations on 11-16-200710:02 AM
SuperfellSuperfell
No, it went out earlier, but we postponed a change until last night to let informatica get the patch out. Informatica were supposed to have contacted all their salesforce.com customers about this issue.
IntegrationsIntegrations

Hey thanks Simon for your help!!!

Our informatica admin went out and did a search on the informatica site with the information you provided and sure enough, there is another patch that needs to be applied.

Tammy

 

 
WesmanWesman
It appears over the past couple of weeks that the endpoint that is being returned by the api during the SSO process has been changing. For a long time it has the following form:

http://na1.salesforce.com/services/Soap/u/8.0

A couple of weeks ago they added a trailing slash:

https://na1.salesforce.com/services/Soap/u/8.0/

Last night this was changed to have even more information appended to the end of the endpoint like so:

https://na3.salesforce.com/services/Soap/u/8.0/471500D3000000006ad

Not sure if this is related to your problem or not but I thought it might be useful information to have on the boards since I haven't seen it anywhere else.
Alex LazicAlex Lazic

The error you are receiving is due to DTM process crashing. You should be able to verify this by checking NT Event Log or Unix core dump directory.

 

There are two possible scenarios

a) Post Winter 08 Session ID issues – The length of SFDC session ID has increased since Winter 08. This caused buffer overflows in the Informatica SFDC connector. This has been resolved by the Informatica with a specific patch (available form Informatica)

 

b) Random crashes with the above patch – In our case, the cashes stopped occurring when we reduced the frequency of the workflows running SFDC connector. If you have to keep workflows running continuously, try to insert timer tasks (say 1 minute) before the session running SFDC connector. This appears to at least reduce the frequency of the crashes (but hardly be the solution). Nevertheless the connector seams to be unstable (at least on Solaris).

 

Kind Regards

 

Alex