The pool manager cache doesn't actually use memcached
Bug #1425726 reported by
Eric Larson
This bug affects 1 person
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Designate |
Fix Released
|
Medium
|
Eric Larson |
Bug Description
In looking into `designate.
This in memory cache could work on a per-node basis. But, if we are going to support this kind of functionality, we should be sure the client is actually threadsafe by adding some locks around the set/delete operations.
Beyond this though, we should be sure that if we have an actual memcached server defined, that server is used. We can also confirm the memcached cilent is threadsafe or uses a thread pool.
Changed in designate: | |
assignee: | nobody → Eric Larson (eric-larson) |
Changed in designate: | |
importance: | Undecided → Medium |
milestone: | none → kilo-3 |
Changed in designate: | |
status: | Fix Committed → Fix Released |
Changed in designate: | |
milestone: | kilo-3 → 2015.1.0 |
To post a comment you must log in.
It looks like I was wrong:
https:/ /github. com/openstack/ designate/ blob/master/ designate/ openstack/ common/ memorycache. py#L44
Should memcache be a requirement or extra_requires?