2024-07-11T03:42:35.556Z | <Jos Collin> If there's a jenkins build failure unrelated to your PR and you repeatedly post retest (without rebase), it would finally give up and succeed. Wow :D |
2024-07-11T04:26:04.641Z | <Jos Collin> @Venky Shankar Any updates on this? |
2024-07-11T04:26:48.987Z | <Venky Shankar> I'm afraid, no. Got involved with a debug session yesterday. |
2024-07-11T04:26:53.407Z | <Venky Shankar> But its in my list. |
2024-07-11T04:26:58.686Z | <Venky Shankar> sorry! |
2024-07-11T05:01:17.704Z | <Jos Collin> @Venky Shankar Any opinion on this^ ? |
2024-07-11T05:01:52.106Z | <Venky Shankar> I will have to check |
2024-07-11T05:01:59.817Z | <Venky Shankar> Haven't seen the throttler implementation in a while |
2024-07-11T05:02:53.001Z | <Venky Shankar> What's the difference b/w the common throttler and rbd-mirror daemon specific throttler? |
2024-07-11T05:04:45.466Z | <Jos Collin> First look, I found SimpleThrottle in common/Throttle.h is easier to use, but should we follow rbd_mirror/Throttler.h? That's my doubt. I've read some benefits mentioned in the PR of rbd throttler. Will check that again. |
2024-07-11T05:05:36.435Z | <Venky Shankar> Try to understand the necessity of rbd-mirror using its own throttler |
2024-07-11T05:05:46.873Z | <Venky Shankar> once that is clear, we can then decide which to chose from |
2024-07-11T05:07:30.775Z | <Jos Collin> 👍 |
2024-07-11T07:14:48.293Z | <Rishabh Dave> what PR was this? never happened to me. |
2024-07-11T08:49:22.385Z | <Dhairya Parmar> @Venky Shankar <https://tracker.ceph.com/issues/66245#note-14> |
2024-07-11T08:55:12.788Z | <Jos Collin> Random ones. Mostly happened in backport PRs. |
2024-07-11T09:01:45.175Z | <Dhairya Parmar> and <https://tracker.ceph.com/issues/66245#note-15> |
2024-07-11T09:18:11.997Z | <Venky Shankar> will have a look later today |
2024-07-11T09:25:26.834Z | <Venky Shankar> give up and succeed - even if it the jenkins test fails, it would report as success? |
2024-07-11T09:32:47.677Z | <Jos Collin> There's no consistency. That's what I meant. |
2024-07-11T10:27:57.952Z | <Rishabh Dave> @Venky Shankar You've reviewed this PR before and it would be nice to have approval from someone who's better acquainted with this component before merging. If possible, PTAL: <https://github.com/ceph/ceph/pull/56118>. |
2024-07-11T10:28:27.627Z | <Rishabh Dave> @Venky Shankar You've reviewed this PR before and it would be nice to have approval from someone who's better acquainted with this component before merging. QA was successfull.
If possible, PTAL: <https://github.com/ceph/ceph/pull/56118>. |
2024-07-11T10:28:31.764Z | <Rishabh Dave> cc @Jos Collin |
2024-07-11T10:30:03.362Z | <Rishabh Dave> @Venky Shankar Same for this PR from @Neeraj Pratap Singh as well - <https://github.com/ceph/ceph/pull/51332>. QA was successful, will be nice to have approval from someone who is better acquainted before merging |
2024-07-11T10:41:21.912Z | <Jos Collin> thanks @Rishabh Dave |
2024-07-11T10:59:52.690Z | <Rishabh Dave> @Jos Collin FYI: deferring merging of your PR that was under testing since changes you made might not be fully tested. For more details, see - <https://github.com/ceph/ceph/pull/56118#pullrequestreview-2171573292>.
cc @Venky Shankar |
2024-07-11T11:01:07.639Z | <Venky Shankar> I don't have any comment on <https://github.com/ceph/ceph/pull/51332>, but its approved and if it passes fs suite, then please seek reviews from core team since its a generic change. |
2024-07-11T11:01:58.763Z | <Rishabh Dave> > I don't have any comment on <https://github.com/ceph/ceph/pull/51332>, but its approved and if it passes fs suite, then please seek reviews from core team since its a generic change.
okay, i'll defer merging this one too and request review from core. thanks! |
2024-07-11T11:02:28.403Z | <Rishabh Dave> but it looks fine to you too, right? |
2024-07-11T11:02:50.584Z | <Venky Shankar> I don't understand the part of the code that has changed so can't say. |
2024-07-11T11:02:57.266Z | <Rishabh Dave> okay |
2024-07-11T11:12:03.480Z | <Jos Collin> checking |
2024-07-11T11:18:15.786Z | <Jos Collin> <https://pulpito.ceph.com/?branch=wip-rishabh-testing-20240701.141150-debug> contains 3 runs. Which one are you referring? |
2024-07-11T11:18:20.990Z | <Jos Collin> @Rishabh Dave |
2024-07-11T11:20:08.020Z | <Jos Collin> @Rishabh Dave I hope <https://pulpito.ceph.com/rishabh-2024-07-02_06:17:07-fs-wip-rishabh-testing-20240701.141150-debug-testing-default-smithi/> is the actual QA you did? |
2024-07-11T11:20:33.283Z | <Jos Collin> @Rishabh Dave I hope <https://pulpito.ceph.com/rishabh-2024-07-02_06:17:07-fs-wip-rishabh-testing-20240701.141150-debug-testing-default-smithi/> is the QA you did? |
2024-07-11T11:33:11.145Z | <Jos Collin> replied on the PR. |
2024-07-11T11:33:22.831Z | <Rishabh Dave> @Jos Collin FYI: deferring merging of your PR that was under testing since changes you made might not be fully tested. For more details, see - <https://github.com/ceph/ceph/pull/56118#pullrequestreview-2171573292>.
cc @Venky Shankar |
2024-07-11T11:33:29.266Z | <Rishabh Dave> @Jos Collin FYI: deferring merging of your PR that was under testing since changes you made might not be fully tested. For more details, see - <https://github.com/ceph/ceph/pull/56118#pullrequestreview-2171573292>.
cc @Venky Shankar |
2024-07-11T11:43:30.357Z | <Rishabh Dave> <https://github.com/ceph/ceph/pull/56118#issuecomment-2222699469> |
2024-07-11T12:18:46.694Z | <Jos Collin> @Rishabh Dave Sorry I was away. Replied on the PR. |
2024-07-11T12:19:01.331Z | <Rishabh Dave> no problem. i launched tests sometime ago |
2024-07-11T12:20:34.795Z | <Jos Collin> You could block these and rerun: |
2024-07-11T12:20:36.234Z | <Jos Collin> • test_mirroring_init_failure
• test_mirroring_init_failure_with_recovery
• test_cephfs_mirror_blocklist
• test_cephfs_mirror_restart_sync_on_blocklist
|
2024-07-11T19:45:14.037Z | <Wes Dillingham> Is the difference between `ceph fs fail <fs>` and `ceph fs set <fs_name> joinable false` only that `fail` causes actives to become standys while setting the joinable to false doesnt make the existing MDS transition away from being active? |