Hi everyone!

Sorry for asking here. This is a question primarily about NFS (which I
plan to use as part of a heartbeat cluster later on), but I've found a
lot of people discussing it here, while I had troubles finding nfs
specific mailing lists, where you can post to everyone, so it would be
great if anyone could help.

My setup: NFS server — Debian Sqeeze (kernel version 2.6.32-5-686),
test clients — same OS, nfs-utils — 1.2.2-4.

I'm having some issues with rpc.statd functionality or I misunderstand
how it is supposed to work. As I've read:
> The first file locking interaction between an  NFS  client  and  server 
> causes  the  NFS lock
> managers on both peers to contact their local NSM service to store 
> information about the
> opposite peer.   On  Linux,  the local lock manager contacts rpc.statd.

What is this locking iteration? If I open a file on NFS share from
client with mcedit (it is writable to the user), I don't get any
entries in /var/lib/nfs/sm on either server or client. These files
don't show up even if I open shared file concurrently from 2 client
machines .

I can't really think of the reason, why may this not be working. Of
what I have found on the topic, everything seems to be all right: rpc
calls are okay in both directions, rpc.statd and rpc.lockd are running
on server and clients, there are no denying rules (/etc/hosts.deny is
empty for now), there are no errors or warnings about any nfs services
in /var/log/daemon.log.

Here is the output of 'rpcinfo -p':
on server
  program vers proto   port
   100000    2   tcp    111  portmapper
   100000    2   udp    111  portmapper
   100024    1   udp  41579  status
   100024    1   tcp  48859  status
   100021    1   udp  33292  nlockmgr
   100021    3   udp  33292  nlockmgr
   100021    4   udp  33292  nlockmgr
   100021    1   tcp  60197  nlockmgr
   100021    3   tcp  60197  nlockmgr
   100021    4   tcp  60197  nlockmgr
   100003    2   tcp   2049  nfs
   100003    3   tcp   2049  nfs
   100003    4   tcp   2049  nfs
   100003    2   udp   2049  nfs
   100003    3   udp   2049  nfs
   100003    4   udp   2049  nfs
   100005    1   udp  57773  mountd
   100005    1   tcp  58996  mountd
   100005    2   udp  57773  mountd
   100005    2   tcp  58996  mountd
   100005    3   udp  57773  mountd
   100005    3   tcp  58996  mountd

on clients
program vers proto   port
   100000    2   tcp    111  portmapper
   100000    2   udp    111  portmapper
   100024    1   udp  55919  status
   100024    1   tcp  42717  status
   100021    1   udp  59095  nlockmgr
   100021    3   udp  59095  nlockmgr
   100021    4   udp  59095  nlockmgr
   100021    1   tcp  51318  nlockmgr
   100021    3   tcp  51318  nlockmgr
   100021    4   tcp  51318  nlockmgr

The same is true if I ask about rpc services remotely 'rpcinfo -p hostname/IP'.

Please, explain me, when files in /var/lib/nfs/sm have to occur (what
is actually nfs file locking) or is there anything else I need to get
it working? Thanks a lot!
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems

Reply via email to