Dear,

As the issue occured, we did not see any error message in the logs of 
https://[OUR_JENKINS_INSTANCE]/computer/[OUR_SLAVE_SERVER]/log

On which class would you advice us to add logging ?

What would be the link between the fact that a connection issue has perhaps 
occured between the master and the slave, and the fact that the job has 
been triggered at an unexpected time ?

Best regards



Le lundi 30 novembre 2015 09:21:59 UTC+1, Victor Martinez a écrit :
>
> Hi,
> If you connect those slaves via ssh then can you review your networking 
> configuration and verify no interruptions happened when the build ran?
>
> You can add some verbose output based on the logs:
> - https://wiki.jenkins-ci.org/display/JENKINS/Logging
>
> SSH slave/master connections within Jenkins retries up to 5 times to get 
> connected afaik.
>
> Cheers
>
>
> On Monday, 30 November 2015 08:58:32 UTC+1, [email protected] wrote:
>>
>> Dear,
>>
>>
>> We are currently using Jenkins 1.596.3 and we have faced an issue with a 
>> job which was not started at the correct timestamp by the timer.
>>
>>
>> The job is configured to be executed automatically on a daily basis at 
>> 7:30pm CEST.
>>
>>
>> The job execution has been correctly triggered and the build has started 
>> at 7:30pm CEST, as expected.
>>
>>
>> However, the logs of the build show the following error message:
>>
>>
>> Looks like the node went offline during the build. Check the slave log for 
>> the details. 
>> FATAL: channel is already closed
>>
>>  It seems that there has been a communication issue between our Jenkins 
>> master (hosted on a VM) and our Jenkins slave (a physical server)
>>
>>
>> Therefore, the job has failed.
>>
>> At that time, we did not face any hardware issue on the servers.
>>
>>
>> In the jenkins logs, except if I missed it, I did not find any reference 
>> of this communication issue, therefore, I cannot identify the root cause of 
>> this communication issue.
>>
>>
>> Could you please tell if there is a possibility to get more information ?
>>
>>
>> Later during the night (at 00:15am CEST), we have done a restart of the 
>> Jenkins instance and the job has been executed at 00:30am CEST by the 
>> trigger.
>> This execution was not expected because the job is not configured to be 
>> executed at 00:30am.
>>
>>
>> Could you please if this execution is a consequence of the communication 
>> issue faced before ?
>>
>> Could you please describe the process when jobs are aborted for such 
>> issues ? Are they automatically restarted by Jenkins after its next restart 
>> ?
>>
>>
>> FYI, this request relates to 
>> https://issues.jenkins-ci.org/browse/JENKINS-31230 which has been 
>> rejected for the following reason:
>>
>>
>> *Daniel Beck 
>> <https://issues.jenkins-ci.org/secure/ViewProfile.jspa?name=danielbeck> 
>> added a comment - 17/Nov/15 12:54 PM* 
>>
>> *This is an issue tracker, not a support site. Furthermore, 1.596.x and 
>> 1.609.x are no longer supported, you need to show a bug exists on 1.625.x 
>> to get attention in this tracker.*
>>
>> *To get help, ask the jenkinsci-users mailing list, or in #jenkins on 
>> Freenode.*
>>
>>
>> Best regards
>>
>

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-users/4dda2743-3664-415b-bb2a-c30eb12745fc%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to