What exactly is the retry methodology for JOBFs whenever there is a network problem?

Discussion created by Carsten_Schmitz on Feb 2, 2018
Latest reply on Feb 2, 2018 by Carsten_Schmitz

Let's say there's a network outage during a JOBF.

Is the Automic agent retrying the communication, is it somehow fault-tolerant to network issues? Is it supposed to be? Or is it accepted that it simply aborts on the first failure to transmit?

I ask because I have repeatedly seen message "U2000131 - Broken Pipe". This may well be a network error, yes, but still the agent immediately fails the JOBF, which leads to repeated complaints from our users.

Also, what is your take on the general issue? Would you rather have the agent retry to complete the JOBF at a later time? Would you rather have it fail outright? It both has pros and cons I recon: Retrying is nice, but if a file arrives too late because of retries, this might possibly disrupt time-dependant workflows also.