linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* sockets stuck in FIN_WAIT2 and CLOSE_WAIT state in 2.6.10
@ 2005-01-10  2:11 Marc Lehmann
  0 siblings, 0 replies; only message in thread
From: Marc Lehmann @ 2005-01-10  2:11 UTC (permalink / raw)
  To: linux-kernel

After upgrading to linux-2.6.10 (on 2004-12-26), my rsh processes start to
hang despite the server side having closed the connection.

Inspection showed that one of the two rsh processes exited:

   16011 ?        SN     0:00      0     6  1717   600  0.0 rsh ruth cd xmltv && ./upd
   16012 ?        ZN     0:00      0     0     0     0  0.0 [rsh] <defunct>

lsof shows both sockets:

   rsh     16011 root    3u  IPv4 6454542          TCP doom:1014->ruth:shell (FIN_WAIT2)
   rsh     16011 root    5u  IPv4 6454545          TCP doom:1013->ruth:1023 (CLOSE_WAIT)

and the remaining rsh process hangs in select on the FIN_WAIT2 socket:

   select(6, [3], NULL, NULL, NULL

Some of the hung processes are stuck for 14 days now.

I did not have the same problem with 2.6.10-rc1, or 2.6.8.1, which I used
before.

This is on a x86 SMP kernel using Debian GNU/Linux. If this problem is
already known, my apologies, I couldn't access a kernel list archive. If
you want more info, feel free to contact me.

-- 
                The choice of a
      -----==-     _GNU_
      ----==-- _       generation     Marc Lehmann
      ---==---(_)__  __ ____  __      pcg@goof.com
      --==---/ / _ \/ // /\ \/ /      http://schmorp.de/
      -=====/_/_//_/\_,_/ /_/\_\      XX11-RIPE

^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2005-01-10  2:11 UTC | newest]

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2005-01-10  2:11 sockets stuck in FIN_WAIT2 and CLOSE_WAIT state in 2.6.10 Marc Lehmann

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).