-
Type: Task
-
Resolution: Done
-
Priority: Critical - P2
-
Affects Version/s: 1.6.1
-
Component/s: None
-
Environment:Ubuntu 11.04, Mongod/Mongos/Config 2.0.4rc1 - Two shard cluster with 2 Node + Aribiter replica sets in each shard, Mongoid 2.4.6, Heroku Cedar Stack, Unicorn, Sidekiq
When trying to use connection pooling in the 1.6.1 driver, the user is getting the following message:
"could not obtain connection within 5.0 seconds. The max pool size is currently 1; consider increasing the pool size or timeout."
Here is the current mongoid.yml file:
defaults: &defaults
allow_dynamic_fields: false
max_retries_on_connection_failure: 20
use_activesupport_time_zone: true
identity_map_enabled: true
pool_size: 35
pool_timeout: 10.0
The current connection count is 676 and seems to stay around that value, so something appears to be working. However, some jobs continue to fail with the above error.
Gist of the error:
https://gist.github.com/2c1aed9492bc5829e233
Curious about this ticket: http://stackoverflow.com/questions/7583439/rails-3-1-thread-safety-broken
Noted that the app is not using threadsafe!