Bad default values for workers for some services
Bug #1626126 reported by
Brent Eagles
This bug affects 1 person
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
tripleo |
Fix Released
|
Medium
|
Brent Eagles |
Bug Description
Note: a patch for this had been up for awhile but didn't have a bug reported for it. It's a performance issue for glance in tripleo.
The default value for Glance API workers is 0, resulting in a single process for dealing with API requests intended for testing and debugging and is unsuitable for production. A better default would be to leave the value unset so that a suitable default can be scaled according to system resources.
Changed in tripleo: | |
status: | New → Confirmed |
importance: | Undecided → Medium |
assignee: | nobody → Brent Eagles (beagles) |
milestone: | none → newton-rc2 |
Changed in tripleo: | |
status: | Confirmed → In Progress |
summary: |
- Bad default value for Glance API Workers + Bad default values for workers for some services |
To post a comment you must log in.
Actually this might extend far beyond the glance api. Current hieradata is:
service_ configs. yaml:glance: :api::workers: 0 configs. yaml:glance: :registry: :workers: 0 configs. yaml:heat: :api::workers: 0 configs. yaml:heat: :api_cfn: :workers: 0 configs. yaml:heat: :api_cloudwatch ::workers: 0 configs. yaml:heat: :engine: :num_engine_ workers: 0 configs. yaml:keystone: :wsgi:: apache: :workers: "%{::processorc ount}" configs. yaml:neutron: :agents: :metadata: :metadata_ workers: 0 configs. yaml:neutron: :server: :api_workers: configs. yaml:neutron: :server: :rpc_workers: configs. yaml:nova: :api::metadata_ workers: 0 configs. yaml:nova: :api::osapi_ compute_ workers: 0 configs. yaml:nova: :conductor: :workers: 0
service_
service_
service_
service_
service_
service_
service_
service_
service_
service_
service_
service_
So unless the services where the worker count is 0 interpret it as something other than single process, this could suck.
On my 2 CPU controller node I have:
1 process for glance API ns-meta- data-proxy replicator, account-reaper, account-replicator, container-auditor, object-replicator, account-auditor, proxy-server, object-updater and container-updater
1 process for glance registry
1 process each for neutron agents (though I think that's what it should be)
1 process for neutron-
1 process for nova scheduler
1 process for nova-consoleauth
1 process for nova-novncproxy
1 process for aodh listener, notifier and evaluator each.
1 process for cinder scheduler
1 process for ceilometer agent-scheduler, agent-notification, collect and polling each
1 process each for swift container-
1 process for gnocchi-metricd
I'll check the puppet to see if any of these are processorcount and work my way backwards.