Search results

Hmmm very stange... I invite you to configure remote logging (for having fresher info if reboots occurs again). And maybe to subscribe official support from Proxmox. It happened only one time ?

Ok. What did you do as special parameters or installs on the servers that doesn't come from PVE ISO installer ? Did the server already worked on an other OS with any problem ? Did you look at hardware logs for over-temperature or manual hard reset ?

Ok. Can you be more specific on your problem please ?

I understand but it is very difficult to put a diagnostic on this. I saw no evidence in the logs. Maybe somebody will have an idea ? What is the hardware of your servers ?

There is no evidence of quorum loss. Corosync does not complains. This there is no a HA protection mecanism involved. Does it happens regularly ?

Oh yes, indeed. thank you for this clarification.

Hello. I think offline is not a concern. The installer uses ISO packages. Personally, I use the ability to fetch the toml configuration by an URL, very practical. Did you read all the doc ? https://pve.proxmox.com/wiki/Automated_Installation

Hello. How many disk do you have per server ? Is there only one 10Gb network for Ceph ? Did you made a comparative test with an other OS or on the host ?

Maybe you are in the case "it's not a bug, it's a feature" :-) If you use HA, be sure to understand the STONITH mecanism. A visual summary : https://www.virtualizationhowto.com/wp-content/smush-webp/2026/02/proxmox-cluster-state-quorum-and-node-majority-1.png.webp Look at Proxmox Cluster State...

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 ?