Agent service is shutting down by itself

[James Calalo]

X 20140706 11:04:04 <m:257294336:195194096:o:FTP Task.tidal-vlrpcmsapp01a-1.ES_vlrdtesmaster01_1>Ping interval: 900000 millis has been exceeded

. 20140706 11:04:04 <m:257294336:195194096:o:FTP Task.tidal-vlrpcmsapp01a-1.ES_vlrdtesmaster01_1>Service Problem: Ping interval exceeded by service

X 20140706 11:04:04 <m:257294336:195194096:o:FTP Task.tidal-vlrpcmsapp01a-1.ES_vlrdtesmaster01_1>Lost service thread, Agent shutting down!

. 20140706 11:04:04 <m:257294336:195194096:o:tidal-vlrpcmsapp01a-1.ES_vlrdtesmaster01_1>Shutting down: FTP Task

. 20140706 11:04:04 <m:257294336:193158752:o:tidal-vlrpcmsapp01a-1.ES_vlrdtesmaster01_1>FTP Task is shutdown

X 20140706 11:04:04 <m:257294336:193158752:o:File Monitor.tidal-vlrpcmsapp01a-1.ES_vlrdtesmaster01_1>Ping interval: 900000 millis has been exceeded

. 20140706 11:04:04 <m:257294336:193158752:o:File Monitor.tidal-vlrpcmsapp01a-1.ES_vlrdtesmaster01_1>Service Problem: Ping interval exceeded by service

X 20140706 11:04:04 <m:257294336:193158752:o:File Monitor.tidal-vlrpcmsapp01a-1.ES_vlrdtesmaster01_1>Lost service thread, Agent shutting down!

our agent's service tends to stop, prior to its stopping we are getting ftp task logs(as shown above) then the service will shut. It will turn up after you manually start it. Cisco TAC recommended to configure the tagent.ini file with svcthrdtimeout=2.. This is a recurring issue with no definite interval it will appear when you least expect it.

Just want to know some of you encountered it and what were the actions you did?

[Joe Fletcher]

You might want to check to see if there is anything else sharing the tcp port. I've seen instances where agents have been shut down by commands destined for another piece of software. Cheers

[James Calalo]

Thanks for the feedback guys unfortunately svc thread didnt worked for our TES

[Todd Clayton]

We had an issue like this last January with an agent that seemed to be failing during periods of heavy network usage. It was sitting on a blade "next" to the backup server. The hypothesis was that this network congestion was causing the agent to timeout.

Making the svcthrdtimeout change was recommended to us as well, and seems to have cleared up the issue.

[Hemnath Anbhazhagan]

We faced this issue, and found heap error "java.lang.OutOfMemoryError: Java heap space" in the log file, So increased heap min and max size by defining minmem and maxmem variables.