I have a TrueNAS CORE server set up with an NFS share which stores my disk image files (.qcow2). But I am having issues with super slow read / write speeds. Essentially, when I start up a group of VMs, they all load up slowly and I thought maybe it was to much for the network card, switch, or even the TrueNAS box, but come to find out, not even 10mb/s of data is being pulled at a single time. I tested the interfaces and everything works fine, with expected speeds of 100mb/s. For some reason, the disk pool is the only one that is super slow on NFS. Any suggestions?
TrueNAS NFS Disk Pool Super Slow On All Nodes
- Thread starter NextPancake401
- Start date
They all start at the same time? What kind of disks are you using for that pool? How is the pool laid out? raidz or (multiple) mirror vdevs?
Depending on the pool layout and the kind of disks used (SSD or HDD) a lot of random IO can kill performance.
Depending on the pool layout and the kind of disks used (SSD or HDD) a lot of random IO can kill performance.