Enable/Enforce IO Timeouts for Mounts? about lizardfs HOT 2 OPEN

Hello @BloodBlight, thank you very much for registering the issue. We will discuss it in the standup and will give you an answer as soon as possible.
Sorry, did you say mfschunkserverreadto or mfschunkserverwavereadto. If it is the second:
The default value of mfschunkserverwavereadto works well for most cases, but, for better performance, you can change it to be 5 or 10:
-o mfschunkserverwavereadto=5
We will keep you posted on the progress.

from lizardfs.

I have not added ether option, or done any testing other than past experience.

However, if it helps, here is the line from my fstab:

lfsmount /LizardFS fuse mfsmaster=mfsmaster,allow_other,big_writes,nodev,noatime,mfsdelayedinit,nonempty,cacheexpirationtime=1000,mfsattrcacheto=30,mfswritecachesize=2048 0 0

Historically, LFS has caused IO to hang indefinitely should the minimum required chunk servers become unavailable (either due to outage or performance). Same for the master server.

This hasn't really been a big issue, but was hoping there was a quick fix.

from lizardfs.