Hello Proxmoxers,
I have a three node cluster that is solid other than when rebooting. I haven't seen any reference to the same behaviour in searches but the error messages in the web console are:
Bulk Start VMs and Containers | Error:Unexpected Status
CT 101 - Start | Error: unable to read tail (got 0 bytes)
Environment is latest 8.2.4 and this problem has been happening since 7.x so at least it is consistent.
I'm running a selection of CT and VMs across three mini HP G4 Elitedesk 800s. Each of the nodes is running 24GB or 32GB RAM with two NVME drives. Over provisioned for the current use case of home automation and home related services like Frigate/Plex/Pi-Hole etc. Dual networks per node with one running on a pcie header.
Most of the time, one or more of the three nodes does not come up and stay up after a triggered reboot until I have connected a screen and keyboard and logged in manually. The troubled node will reboot itself constantly getting to the stage of appearing green as a node in the datacentre view and when it tries to launch the CT or VM, it hangs and reboots.
I assume my manual login locally allows some service to complete successfully as there is no additional action I need to take.
As this issue has impacted each of the three min PCs sometimes and not others, I assume this is not a BIOS setting.
Log file that generates in the web view:
https://pastebin.com/embed_js/x347Jei1
Pointers and help identifying the cause would be most appreciated.
I have a three node cluster that is solid other than when rebooting. I haven't seen any reference to the same behaviour in searches but the error messages in the web console are:
Bulk Start VMs and Containers | Error:Unexpected Status
CT 101 - Start | Error: unable to read tail (got 0 bytes)
Environment is latest 8.2.4 and this problem has been happening since 7.x so at least it is consistent.
I'm running a selection of CT and VMs across three mini HP G4 Elitedesk 800s. Each of the nodes is running 24GB or 32GB RAM with two NVME drives. Over provisioned for the current use case of home automation and home related services like Frigate/Plex/Pi-Hole etc. Dual networks per node with one running on a pcie header.
Most of the time, one or more of the three nodes does not come up and stay up after a triggered reboot until I have connected a screen and keyboard and logged in manually. The troubled node will reboot itself constantly getting to the stage of appearing green as a node in the datacentre view and when it tries to launch the CT or VM, it hangs and reboots.
I assume my manual login locally allows some service to complete successfully as there is no additional action I need to take.
As this issue has impacted each of the three min PCs sometimes and not others, I assume this is not a BIOS setting.
Log file that generates in the web view:
https://pastebin.com/embed_js/x347Jei1
Pointers and help identifying the cause would be most appreciated.