performance issue on restart of nova compute when there are a large number of vms
Bug #1279055 reported by
Sreeram Yerrapragada
This bug affects 3 people
| Affects | Status | Importance | Assigned to | Milestone | |
|---|---|---|---|---|---|
| OpenStack Compute (nova) |
Won't Fix
|
High
|
Unassigned | ||
Bug Description
When there are large number of VMs in the cloud, there is a performance issue on restarting of nova compute. Compute server tries to sync states of all the vm and does not respond until sync is complete.
| description: | updated |
| Changed in nova: | |
| status: | New → Triaged |
| importance: | Undecided → Wishlist |
| Changed in nova: | |
| assignee: | nobody → Aaron Rosen (arosen) |
Revision history for this message
| Aaron Rosen (arosen) wrote | #1 |
| Changed in nova: | |
| importance: | Wishlist → High |
| tags: | added: network |
Revision history for this message
| zhu zhu (zhuzhubj) wrote | #2 |
| tags: | removed: vmware |
| summary: |
- vmware: performance issue on restart of nova compute when there are a - large number of vms + performance issue on restart of nova compute when there are a large + number of vms |
Revision history for this message
| moorryan (moorryan) wrote | #3 |
Revision history for this message
| Matt Riedemann (mriedem) wrote | #4 |
| Changed in nova: | |
| assignee: | Aaron Rosen (arosen) → nobody |
Revision history for this message
| Sean Dague (sdague) wrote | #5 |
| Changed in nova: | |
| status: | Triaged → Won't Fix |
To post a comment you must log in.
This bug isn't just specific to vmware this occurs on kvm as well. The issue is that when nova starts it queries neutron for the nw information of each instance running on the compute node one at a time. We actually don't need to query neutron for this and instead we should just fetch it from the instance_info_cache table as the nw_info is there.