Closed Bug 637298 Opened 13 years ago Closed 13 years ago

the memcache lib needs to un-blacklist dead nodes faster

Categories

(Cloud Services Graveyard :: Server: Sync, defect)

Product:
Cloud Services Graveyard
Component:
Server: Sync
Platform:
x86_64
Linux
Type:
defect
Priority:
Not set
Severity:
normal

Tracking

(Not tracked)

Status:
RESOLVED WONTFIX

People

(Reporter: tarek, Assigned: tarek)

References

Details

Looks like the latest version of the python-memcached lib fails to restore if the Memcached server is restarted and the Python server stays loaded.

Need to investigate on this and fix it.
It restores fine, but there's a 30 seconds lag before a dead node is retried again, which is too long for a simple memcached restarting.

I've asked Sean about this at https://bugs.launchpad.net/python-memcached/+bug/726654

And will patch it in the meantime so our apps can configure this value to 1 s.
Severity: major → normal
Status: NEW → ASSIGNED
Summary: latest python-memcached seem not to restore of failures → the memcache lib needs to un-blacklist dead nodes faster
Well, We also need that patch: https://bugs.launchpad.net/python-memcached/+bug/728359
I have asked Sean if he could look at those patch this week. If he does not have the time I will release a moz-specific RPM for our needs.
We're going to move to pylibmc, so marking this as wontfix
Status: ASSIGNED → RESOLVED
Closed: 13 years ago
Resolution: --- → WONTFIX
Product: Cloud Services → Cloud Services Graveyard
You need to log in before you can comment on or make changes to this bug.