I have a VPS running Centos 6.3 with a fully updated NFS. When I mount the NFS directory from the client, everything works perfectly fine for approximately 3 minutes, then the client hangs attempting to see the directory.
nfs-utils-1.2.3-26.el6.x86_64
service nfs status
rpc.svcgssd is stopped
rpc.mountd (pid 2544) is running...
nfsd (pid 2609 2608 2607 2606 2605 2604 2603 2602) is running...
rpc.rquotad (pid 2540) is running...
cat /etc/exports
/home/user XX.XX.XX.20(rw,async,no_root_squash)
The client is running Centos 5.8. The directory is mounted using
mount x.x.x.6:/home/user /mnt
When everything is working, I get the following on the client:
/usr/sbin/rpcinfo -p X.X.X.6 | grep mountd
100005 1 udp 892 mountd
100005 1 tcp 892 mountd
100005 2 udp 892 mountd
100005 2 tcp 892 mountd
100005 3 udp 892 mountd
100005 3 tcp 892 mountd
When it stops working, rpcinfo just hangs on the client, however running the above command on the server does return data. There are no logs on the NFS Server side that would indicate an issue. On the client side, I see:
cat /var/log/messages
kernel: nfs: server X.X.X.6 not responding, still trying
The client and server are plugged into the same switch, however they are on different networks. The server is a VPS while the client is a dedicated box. SELINUX is in permissive mode on both client and server, and I've turned iptables off on the server to make sure that was not causing an issue.
Any ideas would be helpful - right now I'm having to restart NFS every two minutes in a cron job to keep it semi working.
Thanks