convergence: QueuePool limit of size 5 overflow 10 reached
Bug #1491185 reported by
Angus Salkeld
This bug affects 2 people
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
OpenStack Heat |
Fix Released
|
Medium
|
Anant Patil |
Bug Description
If you have a really large template with a lot of small resources that can quickly made the DB gets hit very hard.
see: http://
u'TimeoutError: QueuePool limit of size 5 overflow 10 reached, connection
timed out, timeout 30\n'
I think limiting the number of resource checks to a number based on the CPU count might be a good idea.
This should also prevent one user from "swamping" the service with jobs.
Changed in heat: | |
importance: | Undecided → Medium |
status: | New → Triaged |
tags: | added: convergence-bugs |
Changed in heat: | |
assignee: | nobody → Rico Lin (rico-lin) |
Changed in heat: | |
assignee: | Rico Lin (rico-lin) → nobody |
Changed in heat: | |
assignee: | nobody → Rico Lin (rico-lin) |
Changed in heat: | |
milestone: | none → mitaka-3 |
To post a comment you must log in.
I think this bug should focus on finding the root cause of the excessive sessions.
For background: http:// lists.openstack .org/pipermail/ openstack- dev/2015- September/ 073587. html stackoverflow. com/questions/ 3360951/ sql-alchemy- connection- time-out
Also see: http://
We are probably leaking sessions, we need to debug that first before jumping to conclusions.