new code uploaded - argus cygwin support

[Carter Bullard]

WoW!!  Finally a day with computational sunshine  ;o)

This is with .threads for the clients?

I think I've fixed John's ratop segfault, but I need to check it
on other machines to be sure.  I'm still having a locking issue
with ratop() when it reads from a file, with .threads, as on
thread gets done reading the file, but other threads are still
working the records, and they get out of sync,  but I think this
is an easy one.

Radium needs a thorough testing, but so far the new (53+)
seems to be doing well on my systems (recovering associations
after they have been down for weeks, is working again)

Now I need to get the filtering for argus() in complete sync
with the clients (they sometimes lag as the code can't be
shared between the two distro's).  The aggregators need some
cleaning up, (occasionally I see very long lived TCP connections
without any user data), and rahisto() needs the new metrics
added to its list of things it can histofy.

If we stay relatively still, and very lucky, we may be able to make
some progress on release!!!!!!

Carter



On Sep 12, 2007, at 11:10 PM, Peter Van Epp wrote:

> 	Still looks good a little way in. OpenBSD is still happy, it doesn't
> complain about the free when HUPed so it seems to be happy with the  
> string
> dup. Ra on the Mac reading a capture file (which used to balloon to  
> over a gig
> of memory) stayed down at the 30 meg level:
>
>
> vanepp    7018   2.2 -0.1    30872   2080  p1  S     7:40PM    
> 0:01.76 /usr/local/bin/ra3 -S 192.75.244.191:560 -n -D4 -w /var/log/ 
> argus/com_argus
> vanepp    7029   0.0 -0.0    27336    360  p2  R+    7:42PM    
> 0:00.00 grep ra3
> test4:~ vanepp$ ps auxwww | grep ra3
> vanepp    7025  100.0 -0.5    37136  11492  p1  R+    7:41PM    
> 0:25.47 ra3 -r com_argus -n
> vanepp    7018   1.1 -0.1    30872   2080  p1  S     7:40PM    
> 0:01.89 /usr/local/bin/ra3 -S 192.75.244.191:560 -n -D4 -w /var/log/ 
> argus/com_argus
> vanepp    7031   0.0 -0.0    18052    288  p2  R+    7:42PM    
> 0:00.00 grep ra3
> test4:~ vanepp$ ps auxwww | grep ra3
> vanepp    7025  100.0 -0.6    38180  13056  p1  R+    7:41PM    
> 0:31.81 ra3 -r com_argus -n
> vanepp    7018   1.3 -0.1    30872   2080  p1  S     7:40PM    
> 0:01.99 /usr/local/bin/ra3 -S 192.75.244.191:560 -n -D4 -w /var/log/ 
> argus/com_argus
> vanepp    7033   0.0 -0.0    27376    420  p2  S+    7:42PM    
> 0:00.00 grep ra3
> test4:~ vanepp$ ps auxwww | grep ra3
> vanepp    7025  100.0 -0.7    40268  15176  p1  R+    7:41PM    
> 0:40.68 ra3 -r com_argus -n
> vanepp    7018   1.1 -0.1    30872   2080  p1  S     7:40PM    
> 0:02.13 /usr/local/bin/ra3 -S 192.75.244.191:560 -n -D4 -w /var/log/ 
> argus/com_argus
> vanepp    7035   0.0 -0.0    27336    356  p2  R+    7:42PM    
> 0:00.00 grep ra3
> test4:~ vanepp$ ps auxwww | grep ra3
> vanepp    7025  100.0 -0.8    41312  16908  p1  R+    7:41PM    
> 0:48.08 ra3 -r com_argus -n
> vanepp    7018   1.0 -0.1    30872   2080  p1  S     7:40PM    
> 0:02.24 /usr/local/bin/ra3 -S 192.75.244.191:560 -n -D4 -w /var/log/ 
> argus/com_argus
> vanepp    7037   0.0 -0.0    27336    348  p2  R+    7:42PM    
> 0:00.00 grep ra3
> test4:~ vanepp$ ps auxwww | grep ra3
> vanepp    7018   2.0 -0.1    30872   2080  p1  S     7:40PM    
> 0:02.36 /usr/local/bin/ra3 -S 192.75.244.191:560 -n -D4 -w /var/log/ 
> argus/com_argus
> vanepp    7039   0.0 -0.0    27336    340  p2  R+    7:42PM    
> 0:00.00 grep ra3
>
> 	argus on the sensor is low (but so is traffic right now the acid test
> will be tomorrow :-)):
>
> ps auxwwww | grep argus
> root     12857 12.5  0.9  40596 37092 ?        SL   19:37   3:29  
> argus -J -P 560 -i eth0 -i eth1 -U 512 -m -F /scratch/argus.conf
>
> as is the ra listening to the argus:
>
> vanepp    7018   1.0 -0.1    30872   2140  p1- S     7:40PM    
> 0:20.11 /usr/local/bin/ra3 -S 192.75.244.191:560 -n -D4 -w /var/log/ 
> argus/com_argus
> vanepp    7092   0.7 -0.0    27336    340  p1  R+    8:06PM    
> 0:00.00 grep ra3
>
> 	 a quick look didn't see any missing or zero time stamps (although a
> more careful look with a script is probably in order). Looks good  
> so far.
>
> Peter Van Epp / Operations and Technical Support
> Simon Fraser University, Burnaby, B.C. Canada
>