dccd 1.1.9 time outs

Sean Rima sean@tcob1.net
Thu Aug 15 17:40:33 UTC 2002


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Thu, 15 Aug 2002, Vernon Schryver muttered drunkenly:

>> ...  If I do a cdcc stats, I get : stats: no answer from tcob1.net
>> (192.168.0.1,6277) after 12029.81 ms although I know it is running,
>> via ps ax. This will usually happen after I connect to the net, maybe
>> with auto ppp up stuff happening the load average goes higher than
>> what dccd likes.
> 
> It is quite unlikely that the load average as in length of the
> scheduler run queue is made so long by mere PPP hassles that dccd
> would be delayed for 12 seconds.  My guess is that routing is the
> problem.  Perhaps the system's IP address changes, and it takes a
> while for the kernel routing table to catch up.  Perhaps the PPP
> network interface is "UP" but not "RUNNING" or vice versa, and not
> passing packets, but the kernel route points to it for talking to
> 192.168.0.1.
> 
> If my guess is right, then it is possible that changing the DCC server
> name in the map file from 192.168.0.1 to 127.0.0.1 would make the
> problem go away.  It is also possible that if I've guessed right that
> `ping 192.168.0.1` would also fail during those 12 seconds.
> 

It also happens at other times, which I cannot tie down, but I if I do a
ps ax|grep dccd the status is D as opposed to the normal S. Now I have
to admit that I cannot help any further but if it does happen again, I
could do a strace -p on dccd's pid

Sean

- -- 
  Sean Rima                                http://www.tcob1.net
  Linux User:      231986          Jabber:   tcobone@jabber.org
  THE VIEWS EXPRESSED HERE ARE NOT NECESSARILY THOSE OF MY WIFE.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.7 (GNU/Linux)
Comment: Use GPG for Secure Mail

iD8DBQE9W+eQHMnSWn2nApQRApwAAJ958nVV8q/xk/lOMJnszCm8Nx0qkwCgkTeG
I1vtloUDADM9PZyD9H9AC/o=
=VHP+
-----END PGP SIGNATURE-----



More information about the DCC mailing list

Contact vjs@rhyolite.com by mail or use the form.