Hello,
I’m using Virtualized TrueNAS Core in Proxmox. I have a problem when I backup specific 2 VMs in Proxmox to TrueNAS pool attached via NFS. When VZDump backup is made in around 30% or less on those VMs TrueNAS crashes, which causes high Proxmox’s load avarage (up to 50) and in result Proxmox also crashes. I don’t have access to TrueNAS even via noVNC.
I use TrueNAS for about a year and I’ve never experienced something like this. When I do backup of LXC Containers or upload something via SMB there is no problem.
TrueNAS has 16GB of ram, two mirrored 4TB disks with 1TB free space left. I don’t use TrueNAS jails or plugins, only NFS, SMB and storage related stuff.
One time TrueNAS sent me an email with alerts: “Device /dev/da1p2 is causing slow I/O on pool data”, “Device /dev/da2p2 is causing slow I/O on pool data.” and “Device /dev/da0p2 is causing slow I/O on pool boot-pool.”. I’ve made long SMART test, which didn’t detect errors.
Please help me how to get more logs what exactly is causing it and how to repair it.
I pass through disks like here pve.proxmox(.)com/wiki/Passthrough_Physical_Disk_to_Virtual_Machine_(VM), so you think that investing in a disk controller card would help?
Thank you so much for your help. Do you think this controller is enough or I should look for something better? aliexpress(.)com/item/1005006533280553.html