>>> UNRECOVERABLE FATAL ERROR <<< Maximum execution time of 30 seconds exceeded /srv/phab/phabricator/src/infrastructure/query/policy/PhabricatorPolicyAwareQuery.php:0 ┻━┻ ︵ ¯\_(ツ)_/¯ ︵ ┻━┻
Description
Description
Details
Details
| Subject | Repo | Branch | Lines +/- | |
|---|---|---|---|---|
| Fix T102142 - hotfix for phabricator Sprint extension. | operations/puppet | production | +4 -5 |
Revisions and Commits
Revisions and Commits
Related Objects
Related Objects
Event Timeline
Comment Actions
I got another error:
Our servers are currently experiencing a technical problem. This is probably temporary and should be fixed soon. Please try again in a few minutes. If you report this error to the Wikimedia System Administrators, please include the details below. Request: GET http://phabricator.wikimedia.org/project/sprint/burn/1113/, from 10.64.0.171 via cp1043 cp1043 ([10.64.0.171]:80), Varnish XID 2120040637 Forwarded for: 2003:5b:ee5c:46ff:8c3c:769f:70c1:ae51, 10.64.0.171 Error: 503, Service Unavailable at Thu, 11 Jun 2015 19:04:08 GMT
Clickable link for convenience: https://phabricator.wikimedia.org/project/sprint/burn/1113/
Comment Actions
My guess: https://phabricator.wikimedia.org/project/sprint/details/1113/ has "Editable By" set to "Custom Policy" (to allow only five users). Hence more complex to check and ending up with a performance issue.
I'm not sure why this restriction was done (@Jdforrester-WMF might remember); what the actual intention of it is (and if the intention is related to the actual restriction ;-), and I'm curious if using the default policy would make this error vanish.
This comment was removed by Christopher.
Comment Actions
The problem is a direct consequence of the new method that was introduced with the latest commit https://gerrit.wikimedia.org/r/#/c/217183/.
In looking at the query analysis plan, it became totally obvious to me that it was retrieving all project column transactions before filtering them for scope!
I just pushed the fix here: https://gerrit.wikimedia.org/r/#/c/217711/2. It adds a where to the transaction query for tasks in the sprint before it executes (not after...).
I really think that this should be hotfixed if possible.
Comment Actions
@mmodell Can you hotfix this 2 line patch https://gerrit.wikimedia.org/r/#/c/217711/2? Until this is merged, burndown charts are effectively broken.
Comment Actions
Change 217845 had a related patch set uploaded (by 20after4):
Fix T102142 - hotfix for phabricator Sprint extension.
Comment Actions
I already forced the pull on the phab server, but the patch still needs to be merged into operations/puppet to make it official. https://phabricator.wikimedia.org/project/sprint/burn/1113/ loads now, still slowly but it does eventually load.
Comment Actions
Thanks everybody for the quick fix!
Confirming that https://phabricator.wikimedia.org/project/sprint/burn/1113/ works for me now.