Closed Bug 567882 Opened 14 years ago Closed 14 years ago

Memcached server errors on support-stage-new

Categories

(mozilla.org Graveyard :: Server Operations, task)

All
Other
task
Not set
blocker

Tracking

(Not tracked)

RESOLVED DUPLICATE of bug 568175

People

(Reporter: jsocol, Assigned: chizu)

Details

In the past 15 minutes, I've seen 5 of this error (and it's the first time I ever recall seeing it, but we had occasional, vaguer issues with the old memcache client):

======
  File "/data/virtualenvs/kitsune/src/django-pylibmc/django_pylibmc/memcached.py", line 57, in add
    return self._cache.add(smart_str(key), value, self._get_memcache_timeout(timeout))

ServerError: error 10 from memcached_add: SERVER ERROR
======

It's not the most verbose or helpful error message--the memcached docs don't actually list the error codes or their meanings. I'm hoping someone can help us figure out what's going on here.
Jeff suggests flushing memcached after the switch to pylibmc but also says he has "no idea" if it will make the error go away.
Severity: major → critical
This is happening quite a lot and making stage-new pretty unusable now.
Assignee: server-ops → thardcastle
I flushed the cache and it seems to have helped. Are you still seeing the error at all?
(In reply to comment #3)
> I flushed the cache and it seems to have helped. Are you still seeing the error
> at all?

I can create new threads, again. Looks like that helped.
Not sure why I'm one of the few (only?) still seeing this, but I'm constantly triggering 500 Internal Service Errors that generate the memcached tracebacks.
I'm still seeing this, and it's blocking me from testing -> blocker (mrz said we can if it truly blocks testing, and it does).
Severity: critical → blocker
This is an application bug.  The app is trying to store more data than memcached allows.
Status: NEW → RESOLVED
Closed: 14 years ago
Resolution: --- → DUPLICATE
Product: mozilla.org → mozilla.org Graveyard
You need to log in before you can comment on or make changes to this bug.