RAID 5 and 6 arrays have a problem known as the "write hole" affecting their consistency after a failure during a disk write such as a crash or power failure. The problem happens when a chunk of RAID protected data known as a stripe is changed on the array. To make the change, the operating system reads the stripe of data, changes the portion of the data requested, recomputes the disk parity for RAID 5 or RAID 6 then rewrites the data to the disks. If a crash or power outage interrupts that process some of the data written to disk will reflect the new content of the stripe while some on other disks will reflect the old content of the stripe. In general the system may be able to detect that there is a problem by rereading the entire stripe and verifying that the parity portion does not match. The system would have no way to verify which portions of the stripe were written with new data and which contain old data so would not be able to properly reconstruct the stripe after a crash.
VMware ESXi: Hosts crash during VM shutdown with PCI passthrough
When using OVMF with a virtual display (without VGA passthrough),you need to set the client resolution in the OVMF menu (which you can reachwith a press of the ESC button during boot), or you have to chooseSPICE as the display type.
Right after starting a VM (game oriented VM with GPU passthro) One of my cores goes to 100% usage. It stays like that even if I shutdown the VM. Sometimes starting the VM is fine, but then upon shutting it down, it causes the whole UNRAID server to crash.I notice I can temporarily "fix" the issue, of one or more of the cores going to 100%, by going to Shares and editing ANY share at all. Even if is just the Share's comment section. Basically, as long as I can hit 'Apply' when editing a share it 'resolves' the issue and the CPU core(s) goes back to normal. Sometimes, more cores do this and eventually leading to unraid crashing and needing a forceful shutdown. All dockers were turned off. VM works fine by the way. This doesn't appear to be the case with the second VM that doesn't have a pass-throu GPU.It may be connected to something that gets 'restarted/cleared' upon editing one of the shares. I've looked around and the cases I've found were not resolved.
I seem to face the same issue. But just with one my VMs. I have a Windows VM (with GPU passthrough), which works very well. I also have a MacOS VM (with a different GPU passthrough), which crashes Unraid upon rebooting from within the VM.
P_CATERR-N means a Processor Catastrophic Error on your server... Sometimes this errors show up during server POST and then go away the next second; so the best advice is to open a TAC case and see if your crash matches the time CATERR error in the logs so we can tell you if that is the real cause of the reboot/shutdown.
Planned to run my ESXi homelab on this gen 7 NUCNAS server needs the whole disk as native -> I need to passthrouh the AHCI to NAS server and run all other homelab VMs on the other M.2 NVMe disk.Now it comes out that it is not possible to passthrough the Intel Corporation Sunraise Point-LP AHCI controller (8086:9d03). I have spent 3 days on this issue and read all the foums:1) added 8086 9d03 d3d0 false into /etc/vmware/passthru.map2) have played with ESXi drivers, disabled, enabled ahci and vmw_ahci3) have tried it on ESXI 6.0U3 and ESXI6.5U14) have tried with AHCI disk connected and without disk (running ESXI on USB)Reading several forums, it looks me that the passthrough is very demanded feature specially for NAS solutions.My conclusion after 3 days is that I have to find another hardware for my ESXi homelab which supports AHCI passthrough. Any suggstions for very low power consumption device, which is similar to NUC and supports ESXI passthrough?
Im englischsprachigen Blog hatte ich den Artikel VMware ESXi: Hosts crash during VM shutdown with PCI passthrough, wo es um ein sehr spezielles Problem beim PCI passthrough ging. Dort hat sich Blog-Leser Pavel gemeldet und folgende Beobachtung als Kommentar gepostet. 2ff7e9595c
Comments