Last modified: 2014-09-22 22:09:27 UTC

Wikimedia Bugzilla is closed!

Wikimedia migrated from Bugzilla to Phabricator. Bug reports are handled in Wikimedia Phabricator.
This static website is read-only and for historical purposes. It is not possible to log in and except for displaying bug reports and their history, links might be broken. See T49301, the corresponding Phabricator task for complete and up-to-date bug report information.
Bug 47301 - [l10n] LocalisationUpdate can incorrectly report/log success
[l10n] LocalisationUpdate can incorrectly report/log success
Status: NEW
Product: Wikimedia
Classification: Unclassified
Deployment systems (Other open bugs)
unspecified
All All
: Normal normal (vote)
: ---
Assigned To: Nobody - You can work on this!
:
Depends on:
Blocks:
  Show dependency treegraph
 
Reported: 2013-04-17 02:32 UTC by MZMcBride
Modified: 2014-09-22 22:09 UTC (History)
10 users (show)

See Also:
Web browser: ---
Mobile Platform: ---
Assignee Huggle Beta Tester: ---


Attachments

Description MZMcBride 2013-04-17 02:32:07 UTC
I have it on good authority that the LocalisationUpdate script can sometimes incorrectly/inaccurately report success (e.g., outputting "!log LocalisationUpdate completed (1.22wmf1) at Wed Apr 17 02:10:22 UTC 2013" to the server admin log) when in reality the localisation update did not run successfully.

The error-checking or reporting logic needs a review. Filing this as a bug so that this issue doesn't get lost.
Comment 1 Nemo 2013-06-13 09:21:11 UTC
(In reply to comment #0)
> I have it on good authority that the LocalisationUpdate script can sometimes
> incorrectly/inaccurately report success (e.g., outputting "!log
> LocalisationUpdate completed (1.22wmf1) at Wed Apr 17 02:10:22 UTC 2013" to
> the
> server admin log) when in reality the localisation update did not run
> successfully.

Of course concrete examples would help understand the issue.
Comment 2 MZMcBride 2013-06-14 01:33:46 UTC
(In reply to comment #1)
> Of course concrete examples would help understand the issue.

Is there an easy way to verify that the localisation update actually took place? I don't think the date of the most recent update is stored anywhere currently.
Comment 3 Nemo 2013-06-14 05:15:50 UTC
(In reply to comment #2)
> (In reply to comment #1)
> > Of course concrete examples would help understand the issue.
> 
> Is there an easy way to verify that the localisation update actually took
> place? 

Yes, the output you mentioned in comment 0. :)

> I don't think the date of the most recent update is stored anywhere
> currently.

I wonder if you mean that Special:Version should also contain the timestamp of CDB files or something.
Comment 4 MZMcBride 2013-06-14 21:33:57 UTC
(In reply to comment #3)
> Yes, the output you mentioned in comment 0. :)

I can tell you that I updated the localisation cache and I can tell the server log that I updated the localisation cache, but that doesn't mean it was actually updated.

> I wonder if you mean that Special:Version should also contain the timestamp
> of CDB files or something.

Something like that, yeah. Though it's possible that the .cdb file could be modified (touched), but not updated.
Comment 5 MZMcBride 2013-06-14 21:36:39 UTC
It would be helpful to track down the specific update script being run daily. It looks like it's not extensions/LocalisationUpdate/update.php.
Comment 6 MZMcBride 2013-06-14 21:44:02 UTC
(In reply to comment #0)
> I have it on good authority that the LocalisationUpdate script can sometimes
> incorrectly/inaccurately report success [...]

This was based on a comment from Tim, probably in -operations or -tech. A transcript may be helpful here.
Comment 7 Gerrit Notification Bot 2013-10-07 10:34:07 UTC
Change 88009 had a related patch set uploaded by Nemo bis:
Add Icinga check for l10nupdate & drop !log-based alerts

https://gerrit.wikimedia.org/r/88009
Comment 8 Andre Klapper 2014-03-15 23:29:28 UTC
Note: Patch mentioned here got a -1 in Gerrit.
Comment 9 Niklas Laxström 2014-03-21 08:59:02 UTC
Moving to Wikimedia, as logging is not in the extension itself.
Comment 10 Greg Grossmeier 2014-09-22 22:09:27 UTC
The patch from Ori (not Nemo as the Gerrit bot thought) stalled out, unfortunately.

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


Navigation
Links