Last modified: 2014-03-10 20:14:44 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 T64484, the corresponding Phabricator task for complete and up-to-date bug report information.
Bug 62484 - intermittent but frequent Login error on beta labs
intermittent but frequent Login error on beta labs
Status: RESOLVED FIXED
Product: Wikimedia
Classification: Unclassified
General/Unknown (Other open bugs)
unspecified
All All
: Unprioritized major (vote)
: ---
Assigned To: Nobody - You can work on this!
: browser-test-bug
Depends on:
Blocks:
  Show dependency treegraph
 
Reported: 2014-03-10 14:54 UTC by Chris McMahon
Modified: 2014-03-10 20:14 UTC (History)
3 users (show)

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


Attachments
no active login attempt (60.38 KB, image/png)
2014-03-10 14:58 UTC, Chris McMahon
Details

Description Chris McMahon 2014-03-10 14:54:13 UTC
click Login button from http://en.wikipedia.beta.wmflabs.org

Login as usual

Get red-box error "Login error
There was an unexpected error logging in. Please try again. If the problem persists, it may be because you have cookies disabled, and you should check that they are enabled in your browser settings. "

No records appear in the fatals log when this happens. 

After a brief inspection, I see this in both automated tests and manual operation, and it happens more frequently manually.
Comment 1 Chris McMahon 2014-03-10 14:58:06 UTC
Also upon clicking BACK and logging in again, user is presented with error msg. 


Central user log in
No active login attempt is in progress for your session.

See screen shot
Comment 2 Chris McMahon 2014-03-10 14:58:42 UTC
Created attachment 14783 [details]
no active login attempt
Comment 3 Chris McMahon 2014-03-10 15:38:09 UTC
I think Antoine was fixing this as I was reporting it.  It had to do with misconfigured user session caching.  Updates to memcache in pmpta and eqiad should solve this.
Comment 4 Antoine "hashar" Musso (WMF) 2014-03-10 20:14:44 UTC
Earlier today I made a change to be merged which enable memcache writes to both pmtpa and eqiad: https://gerrit.wikimedia.org/r/115623

On beta the sessions are stored in Memcached, unlike production which use a redis backend.  The sessions were only stored on one DC which caused much havoc.

I believe I fixed it with https://gerrit.wikimedia.org/r/#/c/117867/ which made the sessions to be written to both data centers.

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


Navigation
Links