On Thu, 2007-03-15 at 11:17 -0700, Jay Chandler wrote:
Running the nightly from two days ago.
The Dovecot log shows no errors, but I keep getting login errors for a few users.
top shows:
71486 user1 1 96 0 4552K 2408K lockd 0 0:00 0.00% imap 11170 user2 1 96 0 4440K 2480K lockd 0 0:00 0.00% imap
And a ps aux shows that these processes have been open for hours. If I move the mbox that contains their mail, the same error still shows up.
The mboxes reside on our NetApp which is NFS mounted-- does anyone have any wisdom regarding this? As best I can tell this hasn't been a problem before the last few RCs, though I could be mistaken on this.
What does "lockd" mean there? It's waiting for a lock? What does tracing the process say it's doing? http://wiki.dovecot.org/Debugging/ProcessTracing