Remi
Thanks for looking, I can no longer replicate the problem either! I checked out the logs from last week but there was nothing that looked a likely issue.
Robert
Remi Duraffort remi.duraffort@linaro.org writes:
Hello Robert,
I can't reproduce any error while adding callbacks in health-checks.
Do you have any error message in the logs? (/var/logs/lava-server/*.log)
2018-04-13 17:06 GMT+02:00 Robert Marshall robert.marshall@codethink.co.uk:
I should have added that when I ran the test using lava-tool it seemed to consistently work (that is the version labelled as non-working) the difference occurs when run as part of a health check.
Robert Marshall robert.marshall@codethink.co.uk writes:
Here they are
LAVA version 2017.7-1~bpo9+1
Robert
Remi Duraffort remi.duraffort@linaro.org writes:
Hello Robert,
can you attach the two full job definitions?
Thanks
2018-04-13 16:39 GMT+02:00 Robert Marshall robert.marshall@codethink.co.uk:
I had a notification section as follows
notify: criteria: status: complete verbosity: verbose callback: url: http://localhost:8080/scheduler/job/%7BID%7D method: POST token: mytoken content-type: json dataset: results recipients:
- to: method: email email: address@site.org
largely copying that from other examples as experiments. I've been seeing as I changed other parts of the health check that it was failing
- timing out when booting the kernel. If I remove the callback section
so
notify: criteria: status: complete verbosity: verbose recipients:
- to: method: email email: address@site.org
then the health check consistently works. Is there some side effect going on here - or is my callback section causing the error, that localhost url should be ok. I'd expect it to always fail if there's a problem with it?
Robert _______________________________________________ Lava-users mailing list Lava-users@lists.linaro.org https://lists.linaro.org/mailman/listinfo/lava-users
Lava-users mailing list Lava-users@lists.linaro.org https://lists.linaro.org/mailman/listinfo/lava-users