| Summary: | The TCF HIPP job's "hourly git ping" does not seem to work | ||
|---|---|---|---|
| Product: | Community | Reporter: | Martin Oberhuber <mober.at+eclipse> |
| Component: | CI-Jenkins | Assignee: | CI Admin Inbox <ci.admin-inbox> |
| Status: | RESOLVED FIXED | QA Contact: | |
| Severity: | normal | ||
| Priority: | P3 | CC: | thanh.ha, webmaster |
| Version: | unspecified | ||
| Target Milestone: | --- | ||
| Hardware: | PC | ||
| OS: | Windows 7 | ||
| Whiteboard: | |||
|
Description
Martin Oberhuber
Are you sure it's not working? I setup a test job to poll every minute as a quick test and it successfully polled the build every minute [1] (I disabled the polling now). I can also see that your tcf-master job ran a build via polling for it's latest #309 build [2]. It says "Started by an SCM change" and ran 1 minutes 15 seconds past the hour. Target Explorer: Introduce an explizit "hidden" attribute for terminal launcher delegates Commit time: 2014-05-22 04:39:25 (EDT) Build time: May 22, 2014 5:01:15 AM Built by: SCM I do see the past several builds before that though which you manually started the builds yourself. Is it possible you're just not waiting long enough? how long do you usually wait between builds? Something I've noticed is the SCM @hourly does not always run immediately at the top of the hour, it seems to be smart enough to shift the timing a bit so that not all jobs suddenly launch at exactly the same time which in my opinion is a good thing. [1] https://hudson.eclipse.org/tcf/job/thanh-test/ [2] https://hudson.eclipse.org/tcf/job/tcf-master/309/ Hi Thanh, thanks for checking. I also noticed that polling seems to work again. Unfortunately the older jobs are gone now. But I know 100% sure that NO job had been scheduled between 17-May and 22-May while I know for sure we've had lots of checkins during that time. I just happened to find out when our downstream build was broken due to a missing dependency. Not sure if anything can be determined in retrospect ? I'll set this to resolved since it seems to be working now. (In reply to Martin Oberhuber from comment #2) > Not sure if anything can be determined in retrospect ? Unfortunately not as far as I can tell. If this happens again reopen this bug and we can keep an eye on it. |