2024-08-06T00:13:20.472Z

<alvaro.soto> are all nodes impacted with your mentioned networking issue? or just one node?

2024-08-06T01:27:24.062Z

<jmguzman> One node is impacted by networking issue, but as OSDs flap.... there is an impact in all the cluster

2024-08-06T01:37:45.783Z

<alvaro.soto> If versions are equal, that rules out software related (ceph)... I'm not sure what do you want to do here, reconfigure the cluster until you fix your network? Then set nodown, noup, norebalance

2024-08-06T01:38:31.768Z

<alvaro.soto> Basically put the cluster in maintenance mode, and hurry up do fix the network issue

2024-08-06T01:39:20.807Z

<alvaro.soto> Safest move will be to move the data out of the failing node, if you have enough space

2024-08-06T01:39:45.113Z

<alvaro.soto> Safest move will be to move the data out of the failing node, if you have enough free space

2024-08-06T02:34:44.288Z

<Samuel Just> @Matthews Jose That's a fairly complicated question.  RADOS actually already has a replica-read mechanism used (at the moment) mainly for reads from rbd clones.  As it happens, there's an outstanding PR <https://github.com/ceph/ceph/pull/56677> improving and clarifying that mechanism.  Reading the commits and commit messages in <https://github.com/ceph/ceph/pull/56677> will probably highlight all of the important touchpoints.

2024-08-06T03:23:55.091Z

<Xinliang Liu> @Adam Kraitman thanks, I accepted the invitation and it works now. 👍
 By the way, are there any rules for the branch naming? I can see many branches start with wip-xx.
And where can I find my built arm64 ceph rpms? I can find the built arm64 container image from <https://quay.ceph.io/repository/ceph-ci/ceph>.
But I can't find the arm64 ceph rpms from <https://3.chacra.ceph.com/r/ceph/ceph-nvmeof-mon-arm64-testin/0adaeecf622b3867ee27e9fabeb939831c4967dc/centos/9/flavors/default/aarch64/>

2024-08-06T06:00:53.732Z

<Deepika Upadhyay> I am not sure what is majorly being used, but Jaeger was using gRPC. I can ask around and get back to you.

2024-08-06T09:54:42.195Z

<cpebble> Hey, I was in here last week with a bug triggering a system crash. We've narrowed it down even more, and i'm starting to think it's a bug in ceph

2024-08-06T09:55:29.145Z

<cpebble> On affected hosts, with some nvme drives(WD Red 700), when `smartctl` is run in a container(tested 18.2.2, trying 18.2.4 atm) the whole system goes down

2024-08-06T09:56:18.579Z

<cpebble> Honestly, where do i even file this? Is the bug in Smartctl, Docker, or Ceph?

2024-08-06T13:13:01.468Z

<Naman Munet> Hi everyone,
Is anyone here running ceph-dev cluster on macos silicon, I am having some issue with my setup where I am using colima based docker setup.
is there any other way to have it up and running?

2024-08-06T13:46:33.030Z

<Rico> @Ernesto Puerta do you have any news ?

2024-08-06T15:11:04.827Z

<John Mulligan> CDS is next week, eh?  Seems kinda like short notice.

2024-08-06T15:15:02.757Z

<Ernesto Puerta> Nope, I tried to remove all problematic Python deps, but sooner or later some indirect dep relies on `cryptography`... My hope was to see some progress in [this PyO3 PR](https://github.com/PyO3/pyo3/pull/4162), but no luck either.

An alternative **temporary workaround** would be to embed a previous working version of the deps together with the ceph-mgr (there was [an attempt to implement the infra in the ceph-mgr to support that](https://github.com/ceph/ceph/pull/47501)).

2024-08-06T15:39:27.237Z

<Rost Khudov> @baum could you pls check [following comment](https://github.com/ceph/ceph/pull/54671#discussion_r1705674118) on one of your recently [merged PR](https://github.com/ceph/ceph/pull/54671)?

2024-08-06T19:14:16.069Z

<Ken Dreyer> What time is <https://tracker.ceph.com/projects/ceph/wiki/CDM_07-AUG-2024>  ? What is the URL to the meeting?

2024-08-06T19:18:15.633Z

<Casey Bodley> 11am eastern in <https://meet.jit.si/ceph-dev-monthly>

2024-08-06T19:19:08.267Z

<Casey Bodley> <https://www.meetup.com/ceph-user-group/events/302555284/> looks like a broken link (cc @Laura Flores)

2024-08-06T19:21:49.464Z

<Laura Flores> Hmm, it looks like a problem with [Meetup.com](http://Meetup.com) itself. I can't fix that, but I'll update the mailing lists with written details if people can't access it.

2024-08-06T19:22:25.291Z

<Laura Flores> It happens if you click on any meetup event.

2024-08-06T21:44:25.224Z

<Laura Flores> The meetup link is back up!