Linux – NFS lock not working properly

linuxnfsredhat

I'm getting the following error on an oracle database.

ORA-27086: unable to lock file - already in use

It tries to lock a file that resides in an NFS mount.

The NFS is on a loopback setup (long story short, I need it this way because of direct nfs) and I traced the error to nlockmgr not going up, even when I restart nfs, nfslock and rpcbind.

Does anybody knows how to get nlockmgr up again?

p08[oracle] () /$ cat /etc/issue
Red Hat Enterprise Linux Server release 6.5 (Santiago)

Here's the exit for rpcinfo -s:

[root@p08 init.d]# rpcinfo -s
   program version(s) netid(s)                         service     owner
    100000  2,3,4     local,udp,tcp,udp6,tcp6          portmapper  superuser
    100011  2,1       tcp,udp                          rquotad     superuser
    100005  3,2,1     tcp6,udp6,tcp,udp                mountd      superuser
    100003  4,3,2     udp6,tcp6,udp,tcp                nfs         superuser
    100227  3,2       udp6,tcp6,udp,tcp                nfs_acl     superuser
    100024  1         tcp6,udp6,tcp,udp                status      29

EDIT:

I found a script executed by the sysadmin after every reboot. After the reboot, nlockmgr is running. After executing this script, it vanishes from rpc.

service xinetd stop
service rpcbind stop
service nfslock stop
service nfs stop
service xinetd start
service rpcbind start
service nfslock start
service nfs start

Best Answer

Check your logs for messages containing locking. If there are problems setting up logging, then it should should be locked.

Is statd running? It is needed for locking to work on NFS versions before 4.

Alternatively, you may try mounting with the option for local lock management. This is not recommended if you have remote clients for the file systems.

Related Topic