About error in GCI Link Inicialization

[J.Calvo]

Can anybody tell us how to solve this problem, and the causes of this?
Thank you !!

[ArndW]

Your question is impossible to answer using the information given; it is like expecting someone to diagnose a car problem by saying "I hear a strange noise sometimes".

Please add some more details to your query: the full error message for a start, what stage is causing the error, if it is sporadic or always happens, whether just one job is affected or more, if the job once worked and now no longer does. Any additional information will make analysis that much easier and get you a quicker response.

[J.Calvo]

Ok sorry, I have been looking for posts of other people talking about this error, and I have confirmed witch this is a common error. I have not found a solution to resolve it and Because of this I have no write anything in particular of our case, but a'll try to answer all of the questions witch you have asked me about it...

In the Log of the Job appear first of all one WARNING:

WARNING:
Volcado_Clients3..InterProcess_1.DSLink206: ds_ipcopen() - call to OpenFileMapping() failed - The system cannot find the file specified.

Next appear a FATAL ERROR:

FATAL ERROR:
Volcado_Clients3..Transformer_7: |Error 11 in GCI Link initialisation.|

And finally another WARNING:

WARNING:
Attempting to Cleanup after ABORT raised in stage Volcado_Clients3..Transformer_7

The fatal error happened in a transformer stage.

A few days ago the job have runed succesfully, but now fails. My colleague who has developed the job tells me that he haven't do any change in the job...

Thanks!

[ArndW]

What stages (name and type) are at the left and right ends of "DSLink206"? There seems to be a runtime problem with the interprocess communication here. Also, if you do a "reset" of the aborted job in the director you will get a log entry "from previous run.." and please post that contents as well.

[J.Calvo]

What stages (name and type) are at the left and right ends of "DSLink206"? There seems to be a runtime problem with the interprocess communication here. Also, if you do a "reset" of the aborted job in ...

Before exist an interproccess .. I have revised the job now, in in the performance options of the interprocess i have found something that i think was cause the problem in ... timeout: my colleague set it to 0, and in buffer size is setted to 32000 or something like that, ... I set int to 2048 and 0 ... and the problem continues, I have next set it to 1024 and a timeout = 600 and the job works fine...

The problem is resolved... but what was the problem???

[chulett]

Neither of those values should ever be set to zero.