[prev in list] [next in list] [prev in thread] [next in thread] 

List:       fwtk-users
Subject:    Re: [fwtk-users] Lots of smap processes on CentOS5/RHEL5
From:       "Michael St. Laurent" <mikes () hartwellcorp ! com>
Date:       2008-04-17 18:19:05
Message-ID: 3BF93070B3D1B047BA7ABF612958950D02CF6121 () hcex ! hartwellcorp ! com
[Download RAW message or body]

> A while back I moved smap and smapd onto a new system running 
> CentOS5 (a
> respin of RHEL5).  When I did this I started seeing a lot of smap
> processes stacking up.  I assume that what is happening is that poorly
> behaved senders are not bothering to end the SMTP transaction cleanly
> and thus the smap processes are hanging around waiting for a signoff
> that is never going to come.  I worked around this problem by 
> running a
> script once a minute that kills any smap process that is more than ten
> minutes old.  Now that I have time to revisit this situation I would
> like to solve it more cleanly.
> 
> Has anyone else run into this problem?

I used gdb to attach to one of the hung processes so as to get a
backtrace, etc.

[root@hcfw1 etc]# gdb smap 8100
GNU gdb Red Hat Linux (6.5-25.el5rh)
Copyright (C) 2006 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you
are
welcome to change it and/or distribute copies of it under certain
conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB.  Type "show warranty" for
details.
This GDB was configured as "i386-redhat-linux-gnu"...smap: No such file
or directory.

Attaching to process 8100
Reading symbols from /usr/local/etc/smap...done.
Using host libthread_db library "/lib/libthread_db.so.1".
0x0806ef4e in __read_nocancel ()
(gdb) bt
#0  0x0806ef4e in __read_nocancel ()
#1  0x08049311 in crlfgets (buf=0xbfd900c8 "", siz=32768,
    fp=<value optimized out>) at smap.c:2720
#2  0x0804ab2c in main (ac=0, av=0xbfd98334) at smap.c:1388
(gdb) list
741             int             sitelen;
742             int             daemonflag = FALSE;
743             int             port_used = -1;
744             char            *fwtk_version   = defined_fwtk_version;
745             char            *bind_addr = NULL;
746
747             int     stop = FALSE;
748             int     rejecting = FALSE;
749
750             if (ac > 0) {
751                     myname = strrchr(*av, DIRC);
752                     if (myname == (char *) NULL)
753                             myname = *av;
754                       else
755                             ++myname;
756             }
757
758             if (atof(fwtk_version) < atof(minimal_fwtk_version))
759                     fwtk_version = minimal_fwtk_version;
760
(gdb)

I've got a timeout value set in the netperm-table but it doesn't seem to
be working.

Does anyone have time to loan me a clue here?

_______________________________________________
Fwtk-users mailing list
Fwtk-users@buoy.com
http://www.buoy.com/mailman/listinfo/fwtk-users
[prev in list] [next in list] [prev in thread] [next in thread]