[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[cobalt-users] POP3 down - port already in use?
- Subject: [cobalt-users] POP3 down - port already in use?
- From: Parker Morse <morse@xxxxxxxxxxx>
- Date: Fri May 2 07:42:00 2003
- List-id: Mailing list for users to share thoughts on Sun Cobalt products. <cobalt-users.list.cobalt.com>
So, this morning I come in and find that POP3 has died on my Qube3. As
best I can tell from the logs and from Active Monitor, it stopped
responding around 10 PM EST last night, give or take half an hour.
Everything else appears to be running properly.
I tried restarting POP in the GUI (uncheck the box, click "Save", recheck
the box, click "Save" again). No good. I also tried restarting inetd from
within SSH: #/etc/rc.d/init.d/inet restart . Also no good.
The output of "netstat -tpa" includes a lot of this:
tcp 1 0 bluebird-eth1.sina:pop3 10.6.18.55:3576
CLOSE_WAIT -
...with 10.6.18.55 being a local client, and the port numbers varying.
There are close to 100 lines of this in netstat! Since there is no process
number shown, I can't close down the processes.
I'm seeing things like this in the log when I attempt to restart POP:
May 2 09:09:51 bluebird inetd[14078]: pop-3/tcp: bind: Address already in
use
When I telnet to the server on port 110, I get a "connected" message, but
no qpopper banner and no response to commands.
Are these earlier connections bound up in CLOSE_WAIT what's choking popd?
The load averages are pretty tiny right now - it just doesn't respond to
POP connections. Client MTAs connect, then time out trying to authenticate.
What else should I try?
I've got my users on webmail now, but I'd love to figure out how to bring
back POP function without rebooting the box.
Thanks,
pjm