nfs works fine after a reboot but as soon as rpcbind is restarted, we cannot get file locks. only solution has been to reboot, afaict. this problem occurs on several different centos 6 servers and I've duplicated it with fresh installs.

server:

client:
*every* time:
- on server boot, nfs is exported fine
- on client boot, nfs mounts are mounted fine
- on client, run file lock test, successful
- on server, issue a rpcbind restart
- on client, run file lock test, failed
- on server, restart all nfs daemons (rpcbind, nfslock, nfs)
- on client, run file lock test, failed
- on server, reboot, wait for up
- on client, run file lock test, successful

for now, we're just resolved to not restart rpcbind but this cannot be expected behavior, right?

Please try nfslock restart with rpcbind, before or after.

nfslock must also be started for both the NFS client and server to function properly.
To start NFS locking, use the following command:

If NFS is set to start at boot, ensure that nfslock also starts by running
If nfslock is not set to on, this implies that you will need to manually run the service nfslock start or restart each time .


nfslock is only needed for NFSv2 and NFSv3.

As you've got Centos v6, go with NFSv4 and you can forget all the 'extra' daemons. See Section 9.1 https://access.redhat.com/knowledge/...de/ch-nfs.html

@ a31amit: I've tried all combinations of restarting those services in every order, all with the same results. nlockmgr will not show in rpcinfo and file locks will be unavailable unless I reboot the server.

@ chrism01: i'm looking into moving those mounts to nfs v4, thanks.