Having a terrible time determining where fault lies in this issue (hw/sw/both??)
I have noticed recently that my VMs (either Debian 12 or Ubuntu 22) would shutdown seemingly randomly. In some cases this would be on initial creation, some would be after creation only after I attempt to perform an activity. There is no error message in the Proxmox UI, and the only indication that this even occurred had to be obtained from journalctl:
Apr 14 09:00:13 dxp480t QEMU[1099916]: kvm: block/block-gen.c:1692: blk_lock_medium: Assertion `!qemu_in_coroutine()' failed.
As evidenced by the above snippet I use this terraform provider: https://registry.terraform.io/providers/Telmate/proxmox/
VMs created under versions RC1 & RC8 exhibit the same behavior. I am currently testing w/o the provider to see if this is the culprit also.
I came across this discussion (https://forum.proxmox.com/threads/sudden-bulk-stop-of-all-vms.139500/page-2) and have performed a microcode update. The machines this behavior is happening on is UGREENAS hardware (480t shown above, 6800pro has same issue)
I have noticed recently that my VMs (either Debian 12 or Ubuntu 22) would shutdown seemingly randomly. In some cases this would be on initial creation, some would be after creation only after I attempt to perform an activity. There is no error message in the Proxmox UI, and the only indication that this even occurred had to be obtained from journalctl:
Apr 14 09:00:13 dxp480t QEMU[1099916]: kvm: block/block-gen.c:1692: blk_lock_medium: Assertion `!qemu_in_coroutine()' failed.
Code:
Apr 14 08:59:51 dxp480t kernel: tap9644i0: entered promiscuous mode
Apr 14 08:59:51 dxp480t kernel: vmbr0: port 2(tap9644i0) entered blocking state
Apr 14 08:59:51 dxp480t kernel: vmbr0: port 2(tap9644i0) entered disabled state
Apr 14 08:59:51 dxp480t kernel: tap9644i0: entered allmulticast mode
Apr 14 08:59:51 dxp480t kernel: atlantic 0000:74:00.0 enp116s0: entered promiscuous mode
Apr 14 08:59:51 dxp480t kernel: vmbr0: port 2(tap9644i0) entered blocking state
Apr 14 08:59:51 dxp480t kernel: vmbr0: port 2(tap9644i0) entered forwarding state
Apr 14 08:59:51 dxp480t pvedaemon[1099902]: VM 9644 started with PID 1099916.
Apr 14 08:59:51 dxp480t pvedaemon[1500]: <automata@pve!terraform> end task UPID:dxp480t:0010C87E:016BD233:67FD06C6:qmstart:96>
Apr 14 08:59:55 dxp480t pvedaemon[1500]: VM 9644 qmp command failed - VM 9644 qmp command 'guest-ping' failed - got timeout
Apr 14 08:59:58 dxp480t pvedaemon[1499]: <root@pam> starting task UPID:dxp480t:0010C8D8:016BD574:67FD06CE:vncproxy:9644:root@>
Apr 14 08:59:58 dxp480t pvedaemon[1099992]: starting vnc proxy UPID:dxp480t:0010C8D8:016BD574:67FD06CE:vncproxy:9644:root@pam:
Apr 14 09:00:03 dxp480t pvedaemon[1500]: VM 9644 qmp command failed - VM 9644 qmp command 'guest-ping' failed - got timeout
Apr 14 09:00:13 dxp480t QEMU[1099916]: kvm: block/block-gen.c:1692: blk_lock_medium: Assertion `!qemu_in_coroutine()' failed.
Apr 14 09:00:13 dxp480t pvedaemon[1499]: <root@pam> end task UPID:dxp480t:0010C8D8:016BD574:67FD06CE:vncproxy:9644:root@pam: >
Apr 14 09:00:13 dxp480t kernel: vmbr0: port 2(tap9644i0) entered disabled state
Apr 14 09:00:13 dxp480t kernel: tap9644i0 (unregistering): left allmulticast mode
Apr 14 09:00:13 dxp480t kernel: vmbr0: port 2(tap9644i0) entered disabled state
Apr 14 09:00:13 dxp480t kernel: atlantic 0000:74:00.0 enp116s0: left promiscuous mode
Apr 14 09:00:14 dxp480t systemd[1]: 9644.scope: Deactivated successfully.As evidenced by the above snippet I use this terraform provider: https://registry.terraform.io/providers/Telmate/proxmox/
VMs created under versions RC1 & RC8 exhibit the same behavior. I am currently testing w/o the provider to see if this is the culprit also.
I came across this discussion (https://forum.proxmox.com/threads/sudden-bulk-stop-of-all-vms.139500/page-2) and have performed a microcode update. The machines this behavior is happening on is UGREENAS hardware (480t shown above, 6800pro has same issue)