Subject: TCP connection stuck in CLOSE_WAIT
To: 'NetBSD/i386 list' <port-i386@NetBSD.ORG>
From: Gunnar Helliesen <gunnar@bitcon.no>
List: port-i386
Date: 04/17/1998 02:45:19
On our FTP archive running 1.3 release we have a mirroring process that
is never able to complete its mirror run.

We use mirror.pl to mirror a site with thousands of small files in
thousands/3 directories (i.e. 3 files in each directory). After a while
(usually a couple of hours), the mirror process stops doing its thing
and just sits there. top(1) shows:

  PID USERNAME PRI NICE  SIZE   RES STATE WAIT     TIME    CPU COMMAND
13370 wsgweb     2    0 5884K 6152K sleep netcon   0:10  0.00% perl

and netstat(1) shows:

tcp       81      0  193.69.226.250.2980    199.26.180.1.21
CLOSE_WAIT

and there it sits for ever. I've RTFS and I understand that my side is
waiting for a close. Shouldn't it time out? I once let the process sit
like this for a week just to check but it never timed out.

Why is this happening anyway? My side is the client and it's clearly not
done fetching files from the other side, the mirror log tells me that.
Is the other side just dropping the connection without sending the
appropriate incantations?

Is there anything I can do on my side?

Gunnar

--
Gunnar Helliesen   | Bergen IT Consult AS  | NetBSD/VAX on a uVAX II
Systems Consultant | Bergen, Norway        | '86 Jaguar Sovereign 4.2
gunnar@bitcon.no   | http://www.bitcon.no/ | '73 Mercedes 280 (240D)