new small memory argus and clients

[Peter Van Epp]

On Mon, Aug 27, 2007 at 10:55:15AM -0400, Carter Bullard wrote:
> Well, this has all the features, so it will use more memory, but it  
> should level off.
> If not, I know where to look for the bug ;o)  How are we now?
> 
> Carter
> 

	In trouble? When I went to look it had seg faulted:

ps auxwwww | grep argus
root     20755  0.0  0.0   3132   832 pts/1    S+   08:02   0:00 grep argus
[1]+  Segmentation fault      (core dumped) argus -JR -P 560 -i eth0 -i eth1 -U 512 -m -F /scratch/argus.conf >/scratch/debug.log 2>&1

Core was generated by `argus -JR -P 560 -i eth0 -i eth1 -U 512 -m -F /scratch/argus.conf '.
Program terminated with signal 11, Segmentation fault.
#0  0x00000400002a0ddc in .__libc_free () from /lib64/power5+/libc.so.6
(gdb) where
#0  0x00000400002a0ddc in .__libc_free () from /lib64/power5+/libc.so.6
#1  0x000000001002bff8 in ArgusFree (buf=0x27ebe00000000) at argus_util.c:1379
#2  0x0000000010011684 in ArgusCopyRecordStruct (rec=0x40008e0c530)
    at ArgusModeler.c:2753
#3  0x000000001001eca8 in ArgusWriteSocket (output=0x1024f480, 
    client=0x10844480, rec=0x40008e0c530) at ArgusUtil.c:1460
#4  0x0000000010023244 in ArgusOutputProcess (arg=0x1024f480)
    at ArgusOutput.c:525
#5  0x000004000010cd9c in .start_thread () from /lib64/power5+/libpthread.so.0
#6  0x0000040000302c2c in .__clone () from /lib64/power5+/libc.so.6
Previous frame inner to this frame (corrupt stack?)

hcids:/scratch # cat debug.log
  ArgusWarning: argus[20599.0000040000026f50]: (null) started
  ArgusWarning: argus[20599.0000040000026f50]: 09 Jul 78 13:21:04.210396 ArgusGetInterfaceStatus: interface eth1 is up
  ArgusWarning: argus[20599.0000040000026f50]: 09 Jul 78 13:21:04.459432 ArgusGetInterfaceStatus: interface eth0 is up
     ArgusInfo: argus[20599.0000040002b98230]: 31 Dec 69 20:33:04.314032 connect from test4.ucs.sfu.ca

/var/log/messages:

Aug 27 07:18:02 hcids kernel: RING: succesfully allocated 0 KB [tot_mem=12664896][order=11]
Aug 27 07:18:02 hcids kernel: RING: allocated 5425 slots [slot_len=1546][tot_mem=8388608]
Aug 27 07:22:37 hcids kernel: RING: succesfully allocated 0 KB [tot_mem=12664896][order=12]
Aug 27 07:22:37 hcids kernel: RING: allocated 10851 slots [slot_len=1546][tot_mem=16777216]
Aug 27 07:22:37 hcids kernel: RING: succesfully allocated 0 KB [tot_mem=12664896][order=12]
Aug 27 07:22:37 hcids kernel: RING: allocated 10851 slots [slot_len=1546][tot_mem=16777216]
Aug 27 07:30:18 hcids syslog-ng[3231]: STATS: dropped 0
Aug 27 07:39:57 hcids zmd: NetworkManagerModule (WARN): Failed to connect to NetworkManager
Aug 27 07:41:46 hcids sshd[20682]: Accepted keyboard-interactive/pam for vanepp from 142.58.1.234 port 60984 ssh2
Aug 27 07:42:33 hcids zmd: Daemon (WARN): Not starting remote web server
hcids:/scratch # 

which looks to be the same place it was going when .memory was defined.
Anything useful I can look at?

Peter Van Epp / Operations and Technical Support 
Simon Fraser University, Burnaby, B.C. Canada