New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

qa/rgw_snaps: move default_idle_timeout config under the client #15400

Merged
merged 1 commit into from Jun 1, 2017

Conversation

Projects
None yet
3 participants
@yehudasa
Member

yehudasa commented Jun 1, 2017

Fixes: http://tracker.ceph.com/issues/20128

Signed-off-by: Yehuda Sadeh yehuda@redhat.com

default_idle_timeout: 3600
client.0: null
client.0:
default_idle_timeout: 3600

This comment has been minimized.

@cbodley

cbodley Jun 1, 2017

Contributor

looks like i broke this in #15184 - default_idle_timeout was specific to fastcgi, so it's no longer being parsed and removed from config. this change should work, but it's probably better to remove the setting completely

it looks like the same issue is present in other upgrade tests:

rados/upgrade/jewel-x-singleton/8-workload/rgw-swift.yaml: default_idle_timeout: 300
upgrade/jewel-x/stress-split/7-final-workload/rgw-swift.yaml: default_idle_timeout: 300
upgrade/kraken-x/stress-split/7-final-workload/rgw-swift.yaml default_idle_timeout: 300

mind removing those too?

This comment has been minimized.

@yehudasa

yehudasa Jun 1, 2017

Member

@cbodley @liewegas I don't mind removing them altogether. I'm not sure we'd see the same issue in those tests, I think the problem we were seeing here was that the field was at the wrong level in the hierarchy.

This comment has been minimized.

@cbodley

cbodley Jun 1, 2017

Contributor

the default_idle_timeout field was global to the rgw task (see https://github.com/ceph/ceph/pull/15184/files#diff-995b04809fcabacc3e3ecfaea903a41aL1128). there was also an idle_timeout that could be used per-client

those global fields are removed from config as they're parsed, resulting in a config whose only keys are clients. that's why it's now being mistakenly interpreted as a remote in this test failure. i'd expect the same thing to happen in the upgrade tests

moving the line under client.0 does work to pass the test - it just won't be used anywhere

@liewegas liewegas changed the title from [DNM] qa/rgw_snaps: move default_idle_timeout config under the client to qa/rgw_snaps: move default_idle_timeout config under the client Jun 1, 2017

@liewegas

This comment has been minimized.

Member

liewegas commented Jun 1, 2017

merging this as is just to fix the breakage!

@liewegas liewegas merged commit b5da257 into ceph:master Jun 1, 2017

3 checks passed

Signed-off-by all commits in this PR are signed
Details
Unmodifed Submodules submodules for project are unmodified
Details
default Build finished.
Details
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment