From nobody@FreeBSD.ORG Fri Nov  5 06:32:29 1999
Return-Path: <nobody@FreeBSD.ORG>
Received: by hub.freebsd.org (Postfix, from userid 32767)
	id A00A214CA4; Fri,  5 Nov 1999 06:32:29 -0800 (PST)
Message-Id: <19991105143229.A00A214CA4@hub.freebsd.org>
Date: Fri,  5 Nov 1999 06:32:29 -0800 (PST)
From: br@itchigo.com
Sender: nobody@FreeBSD.ORG
To: freebsd-gnats-submit@freebsd.org
Subject: TCP connections hangs in FIN_WAIT_2 for >30min
X-Send-Pr-Version: www-1.0

>Number:         14722
>Category:       kern
>Synopsis:       TCP connections hangs in FIN_WAIT_2 for >30min
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    freebsd-bugs
>State:          closed
>Quarter:        
>Keywords:       
>Date-Required:  
>Class:          sw-bug
>Submitter-Id:   current-users
>Arrival-Date:   Fri Nov  5 06:40:01 PST 1999
>Closed-Date:    Sat Mar 2 09:22:34 PST 2002
>Last-Modified:  Sat Mar 02 09:22:51 PST 2002
>Originator:     Bodo Rskamp
>Release:        FreeBSD 3.3-STABLE
>Organization:
Itchigo Communications GmbH
>Environment:
FreeBSD alice.itchigo.com 3.3-STABLE FreeBSD 3.3-STABLE #2: Sat Oct 16 18:15:10 CEST 1999     roof@alice.itchigo.com:/usr/src/sys/compile/ALICE  i386

>Description:
A TCP connection hangs in FIN_WAIT_2 after the server process died.
The remote machine keeps sending TCP packets so that the TCP socket
won't go away. The socket is blocked, and the server cannot restart.

Captured output from tcpdump follows.
A is the local host, where the server died.
B is the remote host, where the connection originated.

15:19:28.851890 A.11110 > B.1720: . ack 1 win 4128 (ttl 246, id 102)
15:19:28.851931 B.1720 > A.11110: . ack 1 win 16680 (DF) (ttl 64, id 59322)
15:20:28.890800 A.11110 > B.1720: . ack 1 win 4128 (ttl 246, id 103)
15:20:28.890830 B.1720 > A.11110: . ack 1 win 16680 (DF) (ttl 64, id 59673)
15:21:28.936971 A.11110 > B.1720: . ack 1 win 4128 (ttl 246, id 104)
15:21:28.936995 B.1720 > A.11110: . ack 1 win 16680 (DF) (ttl 64, id 60213)

The server process died at 14:47.

>How-To-Repeat:

>Fix:


>Release-Note:
>Audit-Trail:
State-Changed-From-To: open->feedback 
State-Changed-By: iedowse 
State-Changed-When: Sun Dec 2 12:06:27 PST 2001 
State-Changed-Why:  

Hi, does this problem still exist? If so, it would be useful if you 
could provide a few details about what FreeBSD is doing wrong. The 
issue of the server refusing to restart can be solved with the 
SO_REUSEADDR socket option. 

http://www.FreeBSD.org/cgi/query-pr.cgi?pr=14722 
State-Changed-From-To: feedback->closed 
State-Changed-By: iedowse 
State-Changed-When: Sat Mar 2 09:22:34 PST 2002 
State-Changed-Why:  

Feedback timeout. 

http://www.FreeBSD.org/cgi/query-pr.cgi?pr=14722 
>Unformatted:
