[ceph-users] ceph orch upgrade and MDS service

Hi, would it be possible to split the upgrade feature of the cephadm orchestrator into the current cluster upgrade and a separate upgrade task for the MDS service(s)? When upgrading the MDS there is a short moment when the active MDS has to be restarted. The CephFS is then down for several s

[ceph-users] Re: ceph orch upgrade and MDS service

Hi, Am 05.08.22 um 12:01 schrieb Dhairya Parmar: So could you tell me exactly the use case of this? I can see you mentioning CephFS going down for a few seconds when MDS restarts, so is that what you want to avert? I want to be able to schedule the MDS restart within a 30 minute time range.

[ceph-users] Re: ceph orch upgrade and MDS service

Am 05.08.22 um 12:20 schrieb Dhairya Parmar: Did you try making use of staggered upgrades functionality? Dang. This is so new that I have not seen the feature yet. Thanks. Regards -- Robert Sander Heinlein Consulting GmbH Sc

[ceph-users] Re: Osd-max-backfills locked to 1000

Hi Kenneth, > Any clues here? Thank you very much! > I assume you are running a Quincy version on your cluster. If so, from the Quincy release, mClock is set to be the default queuing scheduler. It was earlier set to the weighted priority queue scheduler (wpq). To maximize the effectiveness of

[ceph-users] Re: Osd-max-backfills locked to 1000

> > Thank you, that explains indeed a few things! :-) > > Thanks for the feedback! This helps a lot in terms of things to optimize in the mClock profiles. But the underlying problem is that we see iowaits/slowdowns on the clients > while rebalancing. > I added some nvme storage and am moving the d

[ceph-users] rgw: considering deprecation of SSE-KMS integration with OpenStack Barbican

Barbican was the first key management server used for rgw's Server Side Encryption feature. it's integration is documented in https://docs.ceph.com/en/quincy/radosgw/barbican/ we've since added SSE-KMS support for Vault and KMIP, and the SSE-S3 feature (coming soon to quincy) requires Vault our B

[ceph-users] Some odd results while testing disk performance related to write caching

Hello i got a small 3 node ceph cluster and i'm doing some bench marking related to performance with drive write caching. the reason i started was because i wanted to test the SSDs i have for their performance for use as db device for the osds and make sure they are setup as good as i can ge

[ceph-users] Ceph needs your help with defining availability!

Hi everyone, One of the features we are looking into implementing for our upcoming Ceph release (Reef) is the ability to track cluster availability over time. However, the biggest *problem* that we are currently facing is basing our measurement on the *definition of availability* that matches user

[ceph-users] Re: cephfs: num_stray growing without bounds (octopus)

On Fri, Aug 5, 2022 at 9:12 PM Frank Schilder wrote: > Hi Dhairya, > > thanks to pointing me to this tracker. I can try an MDS fail to see if it > clears the stray buckets or if there are still left-overs. Before doing so: > > > Thanks for the logs though. It will help me while writing the patch.