Recent content by ghusson

Ok. I think we will try to work on it with Yann and Damien. Would you like to join us ? We could team work in a visioconference.

Hello bbgeek17 ! Indeed, that's my intention. And firstly re-test and understand the manual server shutdown on one site.

Hello Aaron. Thank you very much for this explanation. You saved us several hours of extensive tests . Now, do you think there is a workaround if this case occurs, like shutting down each server in a datacenter ? Good job for the documentation. Suggestion of amelioration : - tell that a CEPH...

Indeed, the LRM does the job normally. the problems could be for example when : - joining cluster - joining ceph cluster

Nice to know ! But I would fear to break things in Promxox (cluster fencing for exemple).

Ah do you test cluster fencing ? Personally too and without any problem. But molly-guard is only efficient and enabled in SSH interactive mode. Thus it shouldn't interfere with internal Proxmox mechanisms (non interactive mode). BUT : molly-guard is NO enabled when using WEB shell... too sad :-(

Indeed. But this situation should not happen. If a node is fenced, having CEPH out of order is OK for me. The problem on the point you mention is that CEPH OSDs didn't start after cluster rejoin.

And here the case with same cluster partition on the two DCs: ======================================================================== ======================================================================== ======================================================================== RQ : DC1...

Hello, With Damien BROCHARD (Ozerim), we made some extensive CEPH tests in lab. Our running and functional lab : we configured a simulation of 2 DC cluster, with CEPH stretched cluster mode and a 3rd DC of only one PVE as tie breaker (CEPH) / quorum vote (Corosync) Our test : We cut the link...

Hello, With Damien BROCHARD (Ozerim), we made some extensive CEPH tests in lab. As I thought we cannot add a new CEPH monitor in proxmox after stretched cluster mode has been enabled : root@training2-vsrv2:~# pveceph mon destroy 400 not enough arguments pveceph mon destroy <monid>...

Hello. Is your CEPH cluster in stretched cluster mode ?

Best practice is to add a Promxox node which will provide Ceph tie breaker and a Proxmox quorum vote.

Hello, this question is answered in the documentation. https://docs.ceph.com/en/squid/rados/operations/stretch-mode/ : When stretch mode is enabled, PGs will become active only when they peer across CRUSH datacenter``s (or across whichever CRUSH bucket type was specified), assuming both are...

What is the output of : hdparm -z /dev/sde

Hello. Journal is managed by systemd. You can use : journalctl -e