wrong, you miss the point of PLP.
it's like a RAID cache accelerator protected by battery.
PLP ensure data is always written to disk, even if crash or power loss.
for me v266 fixed the hang, I'm unable to hang on my two reproducer in a VM test.
with v240 , I'm easy able reproduce the hang in VM + windows log event id 129 vioscsi reset device with CrystalDiskMark test NVMe Profile 5x8GB Read&Write.
Will update production VM this week-end.
HDD is the bottleneck, even bare metal Win10 is slow, even more when Windows Update active.
switch to SSD is mandatory.
mitigations=off as kernel will help old cpu.
edit: indeed, set discard_granularity is the correct fix.
Thanks @ky41083 for sharing link where is tip for global set discard_granularity.
btw, I wonder if the SSD model has something to do with it.
a side fix, downgrade virtio scsi to version 0.1.208...
you don't need NAT from 161 public ip
if 1 guest use the 161 ip, juste remove ip from vmbr1 interface and set iface vmbr1 inet manual
edit: you lost me, where are others ip in your interfaces ? + put content into CODE TAG
it can't work, only one default gateway is possible in a OS.
do not set additionnal public ip on vmbr1 but directly in your guest VM/CT which has vmbr1.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.