stay socket
FRamsay at castelhq.com
FRamsay at castelhq.com
Mon Mar 31 13:00:17 EST 2003
Actually the socket is in a ESTABLISHED state. And yes it's causing a
problem in that a remote server
program is sending messages to the wrong socket. So there is data in the
recv-q (about 9k) but the
process that created the socket is no longer running. I know what program
opened it and it is not running.
I tried using "netstat -p" to get the PID 'just to be save' but netstat
segfalted when using the -p option.
I have managed to kill the socket in a rather brute force method of pulling
the ethernet cable and cycling the network
sub-subsystem. But it's not the appoach if this occures in the field.
-fjr
Frank Ramsay
Systems Programmer
Castel, Inc
100 Cummings Center
Suite 157h
Beverly, MA 01915
(978) 236 1000 (voice)
(978) 236 1197 (fax)
Email: framsay at castelhq.com
|---------+---------------------------->
| | kclark at CetaceanNe|
| | tworks.com (Kevin|
| | D. Clark) |
| | |
| | 03/31/2003 12:27 |
| | PM |
|---------+---------------------------->
>------------------------------------------------------------------------------------------------------------------------------|
| |
| To: FRamsay at castelhq.com |
| cc: discuss at blu.org |
| Subject: Re: stay socket |
>------------------------------------------------------------------------------------------------------------------------------|
FRamsay at castelhq.com writes:
> I have a socket showing up on netstat for a process that is no longer
> running. Is there any command line way to force this socket to close?
> (short of re-booting).
Very precisely speaking, what are you trying to accomplish by doing
this? Is this actually causing a problem?
Let me guess: this is a TCP socket in the TIME_WAIT state?
--kevin
--
Kevin D. Clark / Cetacean Networks / Portsmouth, N.H. (USA)
cetaceannetworks.com!kclark (GnuPG ID: B280F24E)
alumni.unh.edu!kdc
More information about the Discuss
mailing list