best practices for NFS ?

[nsc]

Hello,

My NFS storage is an DRBD HA NFS "cluster" so i'm looking for the right settings in terms of performance and, above all, resilience in the event of a ‘Primary/Secondary’ switchover.

There is not a lot of information on this page https://pve.proxmox.com/wiki/Storage:_NFS

Here my current "mount" :

192.168.99.20:/srv/reactor-exports/nfs on /mnt/pve/nfs-ha-5to-raid5 type nfs4 (rw,relatime,vers=4.2,rsize=1048576,wsize=1048576,namlen=255,hard,proto=tcp,timeo=600,retrans=2,sec=sys,clientaddr=192.168.99.11,local_lock=none,addr=192.168.99.20)

Do i have to switch first version 3.0 of NFS ?

Thanks

nsc

 

[waltar]

No, v4.2 is best for your case. I hope your 192.168.99.20 is not defined to a host for ha nfs and should be a floating IP which is able to be used by both hosts while just activ on one of.

 

[nsc]

Yes 192.168.99.20 is a floating IP (i have 2 DRBD node 192.168.99.21, 192.168.99.22 and one arbiter - diskless 192.168.99.23). The HA is controlled by reactor, all my hosts can access this NFS share.