setup root logging
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
OpenStack Compute (nova) |
Fix Released
|
Low
|
Alexej Ababilov |
Bug Description
When using _setup_
I'm not sure the pressures that moved things to the current implementation, but the "root" (heh) of the issue seems to be that nova.log.
One fix would be to configure the handlers on the real root logger instead of nova_root, and let modules that are expecting logging to be setup by the app take advantage of a properly configured root logger instead of blasting their messages into NullHandler.
FWIW, if you use a log_config that does the right thing with a properly configured root logger it basically works like you'd expect.
Also it seems like at least in sqlalchemy's case it does some fancy stuff to add stream handler's all over the place if you set echo=True in engine_create. So the keystone middleware thing is the best example I have of this actually being a problem.
Changed in nova: | |
milestone: | none → folsom-1 |
importance: | Undecided → Low |
Changed in nova: | |
status: | Fix Committed → Fix Released |
Changed in nova: | |
milestone: | folsom-1 → 2012.2 |
Fix proposed to branch: master /review. openstack. org/7307
Review: https:/