Bug 1196826 - RFE: A build is marked as failed even though one chroot is still running
Summary: RFE: A build is marked as failed even though one chroot is still running
Keywords:
Status: CLOSED CURRENTRELEASE
Alias: None
Product: Copr
Classification: Community
Component: frontend
Version: unspecified
Hardware: Unspecified
OS: Unspecified
unspecified
unspecified
Target Milestone: ---
Assignee: clime
QA Contact:
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2015-02-26 20:40 UTC by Ryan Lerch
Modified: 2016-05-05 13:06 UTC (History)
3 users (show)

Fixed In Version:
Clone Of:
Environment:
Last Closed: 2016-05-05 13:06:29 UTC
Embargoed:


Attachments (Terms of Use)
screenshot (158.10 KB, image/png)
2015-02-26 20:41 UTC, Ryan Lerch
no flags Details

Description Ryan Lerch 2015-02-26 20:40:41 UTC
If a chroot of a build fails, the entire build is marked as "failed" at the top of the build details page when there is still a chroot in the running state.

See attached screenshot for an example.

Even though the screenshot does not show a logged in version of the page, the controls at the top of the page list the buttons for a failed build (Delete, or Resubmit). So there is no option ot cancel the build anymore even though it is still techincally running.

The build should be in a "running" state until all the chroots are either skipped, failed or success.

Comment 1 Ryan Lerch 2015-02-26 20:41:06 UTC
Created attachment 995764 [details]
screenshot

Comment 2 Raphael Groner 2015-06-07 09:56:15 UTC
+1

Comment 3 Raphael Groner 2015-06-07 09:58:45 UTC
Another option would be to add another column into the builds overview table, showing the number of still running chroot jobs.

Comment 4 clime 2016-03-21 16:32:03 UTC
Fixed by b924147b (https://github.com/fedora-copr/copr/commit/b924147bde8a50696be95f7368ac7dbf02dd2cef). Thanks.

Comment 5 Miroslav Suchý 2016-05-05 13:06:29 UTC
Package including this fix has been deployed on production instance of Copr.


Note You need to log in before you can comment on or make changes to this bug.