Search results

Now that I pretty sure it is watchdog. Basically all previous event also same root cause root@pve1:/var/log# journalctl -u watchdog-mux | grep expired Dec 04 06:10:31 pve1 watchdog-mux[1827]: client watchdog expired - disable watchdog updates Dec 27 05:50:50 pve1 watchdog-mux[1824]: client...

Pretty much watchdog root cause. I had this issue multiple times. it all random, When other node return from poweroff, pve1 reloaded. I also check replicated all completed before Sep 01 15:59:59. Sep 01 15:57:15 pve1 corosync[2987]: [KNET ] rx: host: 2 link: 0 is up Sep 01 15:57:15 pve1...

Hi. Thanks you for your reply Unfortunately in this case, 3 node is running and pve1 suddenly reloaded. Is there a way i can check the log confirm it is reloaded due to fencing? If you see the log, there are 3 node Sep 01 15:57:24 pve1 corosync[2987]: [QUORUM] Sync joined[1]: 3 Nothing...

Pve1 was running pvecm e 1. pve2 and pve3 were shutdown and brought up together. How do check logs PVE1 reloaded due to losing quorum ? I brought up pve2 and pve3 15:57:15 -- pve2 started 15:57:24 -- pve3 started 16:09:07 -- Pve1 lost watchdog and reloaded(Losing 3 HA VM) PVE1 Aug 31...

I have some question regarding fencing and node force reboot which recently cause outage to my cluster. Having 3(let call 1,2,3) node cluster, all same version 8.0.3 node 3 having some issue thus was shutdown. No vm is running node 2 is out for maintenance few days ago node 1 is running...

I edit the datastore.cfg I can see the datastore in PBS. I can create new backup to datastore from Proxmox however i cant see all the previous backup data. Update: I notice datastore use some kind of encoding as it was mount from samba. Rename it seems to fix the issue . I am sharing incase...

My machine has 16GB ram(Virtual Environment 7.2-7). VM1 given 2GB and 6GB ballooning enabled(qemu agent enabled). During midnight VM1 will run backup which consuming all 6GB and mostly cache total used free shared buff/cache available Mem: 5035...

To answer to my own question Create a copy of /proc/cmdline to /etc/kernel/cmdline add to end of the line rootflags=rw,degraded eg. initrd=\EFI\proxmox\5.13.19-2-pve\initrd.img-5.13.19-2-pve BOOT_IMAGE=/boot/vmlinuz-5.13.19-2-pve root=UUID=e115f582-3078-4d22-a8e3-xxxxxxxxxxx ro quiet...

I simulate disk fail on the raid1 btrfs which is also my root for proxmox ve 7.1-7. When one of the disk failed, the system will not boot. It stuck at initramfs, i am unable to mount the system as the initramfs does not have mount.btrfs The system will boot normally with both disk exits...