[Orca-users] Problems with orca-aix-stat.pl on AIX 5.1 ML7

[David Michaels]

I would recommend using the orca-aixtsm-stat.pl script, even if you're 
not using tsm.  I've attached my (working) copy here.  We're running 
oslevel 5.1.0.0 on that machine.  Not sure what ML we're up to though.  
My guess is 5.

As for the perl -- there are only three places where the netstat command 
is invoked, and you can find them by doing this:

grep netstat orca-aixtsm-stat.pl | grep open

I get:

dmichael at npdsa01 <~/orca/my AIX orca stuff><27> grep netstat 
orca-aixtsm-stat.pl | grep open
open IN, "netstat -ni|";
        open IN, "netstat -n -I $interface 1 |";
    open IN, "netstat -an |";

The first command there lists the interfaces on the machine.  The second 
one goes through each interface, and gets current error counters (the 
'1' means "just run once").  And the last one shows all the TCP info 
about the machine, using the 'n' flag to say "don't translate the IPs to 
names" (that slams DNS/NIS, and you really don't need the hostnames, you 
just need the stats really).

None of these commands hang up or result in spiked CPU usage when I run 
it on the AIX server mentioned above.  If your netstat is spiking your 
CPU, then either (a) your collection interval is way too small, or (b) 
your script is different than mine, and it's trying to do something it 
probably shouldn't.  In either case, your "ps -ef | grep netstat" can 
find the offending netstat command, and the arguments it was given, and 
that should help you find out where in the script that netstat is being 
run.  Then it can be tweaked/fixed to make it not do that anymore. ;)

I've attached my copy of the orca-aixtsm-stat.pl script, which I think 
was submitted to the devtree around r380, but I could be mistaken.

--Dragon


daniel.engelsen at caremark.com wrote:

>
>
>We recently updated several of our servers ML7 of AIX 5.1, and we are now
>having problems with the orca-aix-stat.pl script.  The netstat command
>keeps going to the top of the process table in terms of amount of CPU
>consumed.  It is taking up to 50% of the CPU.  We were not having these
>problems with prior versions of maintenance on the servers.  So, I called
>IBM and they are saying that it is a problem with the script.  I do not
>work much with Perl.  So, I was hoping that someone else has run into this
>problem and has already found a fix or work around.
>
>Here is the version of the script that I am running.
>
>(See attached file: orca-aix-stat.pl)
>
>Thanks,
>Dan
>  
>
>------------------------------------------------------------------------
>
>_______________________________________________
>Orca-users mailing list
>Orca-users at orcaware.com
>http://www.orcaware.com/mailman/listinfo/orca-users
>  
>

-- 

	Raytheon 	

	*David P. Michaels*
Senior Multi-Disciplined Engineer II W.H.
NPOESS IS
Platform OS Unix
303.344.6840
720.858.5952 fax
720.521.0561 pager
dragon at raytheon.com <mailto:dragon at raytheon.com> 	*aka "Dragon"*

"I wonder what news is doing..."

 news at newshost <29> ps -fu news      
 news 18624 12367 2 0:00 makehistory 

"News is making history."
      

-------------- next part --------------
An HTML attachment was scrubbed...
URL: </pipermail/orca-users/attachments/20041223/c361142c/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Raytheon.gif
Type: image/gif
Size: 481 bytes
Desc: not available
URL: </pipermail/orca-users/attachments/20041223/c361142c/attachment.gif>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: orca-aixtsm-stat.pl.diff
URL: </pipermail/orca-users/attachments/20041223/c361142c/attachment.ksh>