lkml.org 
[lkml]   [2008]   [Jun]   [4]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [bug] stuck localhost TCP connections, v2.6.26-rc3+
On Wed, 4 Jun 2008, Håkon Løvdal wrote:

> Hi. I have now tested and taken various logs before/under/after (ps,
> netstat, tcpdump, strace, /proc/net/tcp). The logs did not grow as big
> as I feared however they are way too big attach here. Any tips on how to
> distribute?

Usually such logs compress quite well. A best thing might be to put
somewhere (if possible) and http:// reference them.

> I mentioned before that the ssh connection died on the new PC, but
> sometimes it just hangs as well, i.e. the ssh connection is alive but
> there is no data transferred.
> The last test I did hang after just 292MB transferred. In this case pv
> just hangs on the new PC, waiting for data that never arrives. On the
> sending side tar which is producing data have died and the corresponding
> ssh is locked in some uninterruptible sleep:

> //netstat before the hang:
> Proto Recv-Q Send-Q Local Address Foreign Address
> State PID/Program name Timer
> tcp 0 504 new_pc:22 old_pc:39464
> ESTABLISHED 17054/sshd: hlovdal on (0.20/0/0)
>
> // after the hang:
> Proto Recv-Q Send-Q Local Address Foreign Address
> State PID/Program name Timer
> tcp 0 2896 old_pc:22 new_pc:39464
> ESTABLISHED - on (21329441.19/0/0)

??? Are these on the same machine? How come the ip-addresses got reversed
(while ports remain the same)...?!?

...Something could be corrupting more than a little, addresses, reference
to the process, timers all seem mixed up here...

> Many things in /proc/17054/ are in a non-normal state (exe is an
> empty symlink, fd is empty, etc)

...


--
i.
\
 
 \ /
  Last update: 2008-06-04 13:17    [W:0.095 / U:3.168 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site