Restart Rundeck jobs after failure

698 views
Skip to first unread message

David Mando

unread,
Oct 17, 2016, 10:36:53 AM10/17/16
to rundeck-discuss
Hello,

My company recently upgraded to Rundeck as our main scheduling tool. We are having issues with jobs failing and not restarting. Does anyone know how I can accomplish this?

Any help would be appreciated.

Alex Honor

unread,
Oct 17, 2016, 11:11:43 AM10/17/16
to rundeck...@googlegroups.com
Hi David,

Can you attach errors you might be getting either from the job or the service.log?

--
You received this message because you are subscribed to the Google Groups "rundeck-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to rundeck-discuss+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/rundeck-discuss/7766f09f-0ddf-46be-9d84-a93f8fb51162%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.



--

Alex Honor

[SimplifyOps, Inc | a...@simplifyops.com ]

Be sure to comment and vote on Rundeck Feature Development!

David Mando

unread,
Oct 17, 2016, 11:17:33 AM10/17/16
to rundeck-discuss
Hello and thanks for your response. I am asking if there is a feature in Rundeck that allows the jobs to restart after failure. The failures that we are getting are internal, so we are trying to develop a defensive coding strategy against this.


On Monday, October 17, 2016 at 11:11:43 AM UTC-4, Alex Honor wrote:
Hi David,

Can you attach errors you might be getting either from the job or the service.log?
On Mon, Oct 17, 2016 at 11:33 AM, David Mando <dman...@gmail.com> wrote:
Hello,

My company recently upgraded to Rundeck as our main scheduling tool. We are having issues with jobs failing and not restarting. Does anyone know how I can accomplish this?

Any help would be appreciated.

--
You received this message because you are subscribed to the Google Groups "rundeck-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to rundeck-discu...@googlegroups.com.

Alex Honor

unread,
Oct 17, 2016, 11:22:27 AM10/17/16
to rundeck...@googlegroups.com
Jobs do have the ability to retry but you will need to define the jobs to check if they are in a retry loop and perhaps also make sure the steps are designed to run again (eg idempotent). 

To unsubscribe from this group and stop receiving emails from it, send an email to rundeck-discuss+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/rundeck-discuss/f8c8bb17-b598-4a78-95d0-064dca06d633%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

David Mando

unread,
Oct 17, 2016, 12:54:44 PM10/17/16
to rundeck-discuss
Thanks Alex.

Ok, so a different question but relevant...

I want to be able to set the job node/step to retry after a failure. Is this possible?

Thank you very much for all your help.



On Monday, October 17, 2016 at 11:22:27 AM UTC-4, Alex Honor wrote:
Jobs do have the ability to retry but you will need to define the jobs to check if they are in a retry loop and perhaps also make sure the steps are designed to run again (eg idempotent). 
On Mon, Oct 17, 2016 at 12:17 PM, David Mando <dman...@gmail.com> wrote:
Hello and thanks for your response. I am asking if there is a feature in Rundeck that allows the jobs to restart after failure. The failures that we are getting are internal, so we are trying to develop a defensive coding strategy against this.


On Monday, October 17, 2016 at 11:11:43 AM UTC-4, Alex Honor wrote:
Hi David,

Can you attach errors you might be getting either from the job or the service.log?

On Mon, Oct 17, 2016 at 11:33 AM, David Mando <dman...@gmail.com> wrote:
Hello,

My company recently upgraded to Rundeck as our main scheduling tool. We are having issues with jobs failing and not restarting. Does anyone know how I can accomplish this?

Any help would be appreciated.

--
You received this message because you are subscribed to the Google Groups "rundeck-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to rundeck-discu...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/rundeck-discuss/7766f09f-0ddf-46be-9d84-a93f8fb51162%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.



--

Alex Honor

[SimplifyOps, Inc | a....@simplifyops.com ]

Be sure to comment and vote on Rundeck Feature Development!

--
You received this message because you are subscribed to the Google Groups "rundeck-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to rundeck-discu...@googlegroups.com.

Alex Honor

unread,
Oct 17, 2016, 2:09:47 PM10/17/16
to rundeck...@googlegroups.com
At a step level you can define an error handler. For a node failure, you can mark the job to keepgoing which will result in a failed job execution status with the list of failed nodes. In essence, you will want to re run the last execution just for them.

To unsubscribe from this group and stop receiving emails from it, send an email to rundeck-discuss+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/rundeck-discuss/419f2e87-5c95-493e-99bf-4f3b41062790%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

David Mando

unread,
Oct 18, 2016, 12:10:13 PM10/18/16
to rundeck-discuss
Thank you very much for your time.
Reply all
Reply to author
Forward
0 new messages