Just as the summary says. rpc.mountd will often exit leaving no trace of reason why in /var/log/messages. The rest of the nfs server system is still up and running (nfsd, and rquotad).
Here is a bit extracted from my /var/log/messages... --snip-- Nov 15 15:14:43 linux4 nfs: rpc.nfsd startup succeeded Nov 15 15:14:44 linux4 mountd[25368]: dump request from 147.117.133.147 Nov 15 15:15:15 linux4 last message repeated 31 times Nov 15 15:16:16 linux4 last message repeated 61 times Nov 15 15:17:17 linux4 last message repeated 61 times Nov 15 15:18:18 linux4 last message repeated 61 times Nov 15 15:19:19 linux4 last message repeated 61 times Nov 15 15:20:20 linux4 last message repeated 61 times Nov 15 15:21:21 linux4 last message repeated 61 times Nov 15 15:22:22 linux4 last message repeated 61 times Nov 15 15:23:23 linux4 last message repeated 56 times Nov 15 15:24:24 linux4 last message repeated 50 times Nov 15 15:25:25 linux4 last message repeated 46 times Nov 15 15:26:26 linux4 last message repeated 42 times Nov 15 15:26:58 linux4 last message repeated 19 times --snip--
This seems to be frighteningly similar to BUG #7546
assigned to johnsonm
Bug 7024 is being closed because of major changes in NFS code since RHL 6.1 and that other similar bugs have been closed with errata that came out for 6.2.