nv-l
[Top] [All Lists]

RE: [nv-l] snmpCollect

To: "'nv-l@lists.us.ibm.com'" <nv-l@lists.us.ibm.com>
Subject: RE: [nv-l] snmpCollect
From: "Liu, David" <david.liu@eds.com>
Date: Wed, 2 Feb 2005 09:18:45 -0000
Delivery-date: Wed, 02 Feb 2005 09:20:38 +0000
Envelope-to: nv-l-archive@lists.skills-1st.co.uk
Reply-to: nv-l@lists.us.ibm.com
Sender: owner-nv-l@lists.us.ibm.com
Joe,

Yes, the default was 5, and I do have verbose trace turned on.

But from the trace file I compare (50 and 100), I cannot see clear
difference. Maybe I was wrong.

Thx & rgds,
David

-----Original Message-----
From: owner-nv-l@lists.us.ibm.com [mailto:owner-nv-l@lists.us.ibm.com]On
Behalf Of Joe Fernandez
Sent: Wednesday, February 02, 2005 3:38 AM
To: nv-l@lists.us.ibm.com
Subject: RE: [nv-l] snmpCollect


David,

Try the command

snmpCollect ?

which will tell you what the default value of NumberConcurrentSnmp is.

Unless this was a new installation of NetView, your "original" value may 
not be the default.

The statement that you can increase it if you have network capacity and 
NetView system capacity makes sense to me.

My guess is that to increase the number of concurrent requests snmpCollect 
is going to have to maintain more state information or fork off new 
processes. In each case this will need more CPU and memory and the question 
is where does it start hitting limits on your box.

You should be able to see what happens by monitoring the trace file as you 
increase the number.

I suggested the -S switch in my previous email. Did that give you 
configuration information that shed some light?

Do you have verbose tracing on?  (-V and -T switches.)



At 01:33 PM 01-02-05 +0000, you wrote:
>Joe,
>
>Thanks. The original config from our NV was 200 and I changed to 100 and 50
>and back to 100. I read somewhere if the network capacity and NV box (CPU)
>have no problem we can have it increased.
>
>Rgds,
>David
>
>-----Original Message-----
>From: owner-nv-l@lists.us.ibm.com [mailto:owner-nv-l@lists.us.ibm.com]On
>Behalf Of Joe Fernandez
>Sent: Tuesday, February 01, 2005 1:56 PM
>To: nv-l@lists.us.ibm.com
>Subject: RE: [nv-l] snmpCollect
>
>
>At 08:11 AM 31-01-05 +0000, you wrote:
> >Thanks Joe,
> >
> >There are no files in the ../snmpCollect directory (files with 0 length).
> >
> >My NumberConcurrentSnmp is 100 and I made change on number of objects in
>the
> >database (from 10000 ro 20000) only.
> >
> >Regards,
> >David
>
>David,
>
>I believe the default value of NumberConcurrentSnmp is 5, so 100 sounds
>ambitious to me.
>
>Did you increment it in steps and monitor the effect?
>
>I would turn it back down, see if data is being collected, then increment
>it in steps and check.
>
>Try the -S switch also and look at the configuration information that it
>writes to the trace file. This should tell you more about what is
happening.
>
>
>
>Joe Fernandez
>Kardinia Software
>jfernand@kardinia.com
>www.kardinia.com

Joe Fernandez
Kardinia Software
jfernand@kardinia.com
www.kardinia.com


<Prev in Thread] Current Thread [Next in Thread>

Archive operated by Skills 1st Ltd

See also: The NetView Web