NSO Tail-f question: application timeout

[lamt2002]

Hi,

I have a NSO service that moves configuration from one device to another. If there is a lot of configuration to be moved, I have run into an application timeout error during "commit dry-run". The timeout occurs roughly 2 mins after "commit dry-run" is executed. Wondering if anyone has encountered similar situation and if there is a way to change the service execution timeout ... thanks in advance.

admin@ncs% set services xxx

[ok][2018-09-12 22:17:12]

admin@ncs% commit dry-run | details
entering validate phase for running...
validate: run pre-trans-lock service callbacks...
validate: run transforms and transaction hooks... ok [0.02 sec]
validate: pre-trans-lock service callbacks done [0.02 sec]
validate: grabbing transaction lock... ok [0.00 sec]
validate: run transforms and transaction hooks... error [121.52 sec]
Aborted: application timeout
[error][2018-09-12 22:19:21]

admin@server1:/var/log/ncs$ tac ncs.log | less
<CRIT> 12-Sep-2018::22:19:21.912 server1 ncs[1391]: - Daemon ncs-dp-2273-serviceXXX:main timed out

admin@server1:/var/log/ncs$ tac devel.log | less
<ERR> 12-Sep-2018::22:19:21.914 server1 ncs[1391]: devel-c service_create error {external_timeout, ""} for callpoint 'serviceXXX-servicepoint' path /ncs:services/serviceXXX{test1}
<ERR> 12-Sep-2018::22:19:21.911 server1 ncs[1391]: devel-c Worker socket query timed out daemon 'ncs-dp-2273-serviceXXX:main' id 49

Thanks so much.

Tim

[rogaglia]

Hi,

Hard to say without looking at your code but are you calling an external action from your service create() code? maybe opening another maapi session?

Regards,

Roque

[lamt2002 · Accepted solution]

Thank you Roque for your reply.

Thanks to the suggestion by Cisco TAC to increase the following timer in NSO (default is 120 sec), the service commit dry-run is now working ... 

admin@ncs% set java-vm service-transaction-timeout

(<int>) (120):

Thanks so much.