Re: negative tcp_tw_count and other TIME_WAIT weirdness?

["David S. Miller" <davem@redhat.com>]

   From: John Salmon <jsalmon@thesalmons.org>
   Date: Mon, 30 Jun 2003 17:25:16 -0700

   I have several fairly busy servers reporting a negative value
   for tcp_tw_count.

 I have a sneaking suspicion that this patch (already in 2.4.22-preX)
 will fix your problem.

# This is a BitKeeper generated patch for the following project:
# Project Name: Linux kernel tree
# This patch format is intended for GNU patch command version 2.5 or higher.
# This patch includes the following deltas:
#	           ChangeSet	1.930.114.22 -> 1.930.114.23
#	net/ipv4/tcp_minisocks.c	1.13    -> 1.14   
#
# The following is the BitKeeper ChangeSet Log
# --------------------------------------------
# 03/05/07	olof@austin.ibm.com	1.930.114.23
# [TCP]: tcp_twkill leaves death row list in inconsistent state over tcp_timewait_kill.
# --------------------------------------------
#
diff -Nru a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
--- a/net/ipv4/tcp_minisocks.c	Tue Jul  1 01:25:26 2003
+++ b/net/ipv4/tcp_minisocks.c	Tue Jul  1 01:25:26 2003
@@ -447,6 +447,8 @@
 
 	while((tw = tcp_tw_death_row[tcp_tw_death_row_slot]) != NULL) {
 		tcp_tw_death_row[tcp_tw_death_row_slot] = tw->next_death;
+		if (tw->next_death)
+			tw->next_death->pprev_death = tw->pprev_death;
 		tw->pprev_death = NULL;
 		spin_unlock(&tw_death_lock);