[rtg] Traffic is wrapping

Tue Apr 8 00:28:26 EDT 2008

We're polling a few hundred devices and over 4000 total ports. The only
performance / scalability issues we've run into thus far is a memory leak on
the rehashing of the targets. 

We're running on a C2D 6550, 4gb ram. 

We actually 'roll up' the data out of the main RTG database and pull it into
separate databases so that eliminates the 'growth' problems of the rtg
tables and the requirements for heavy repair / optimizes. 

We also then summarize the data into hourly figures, daily figures in / out,
etc. It helps keep down table size for highly 'active' data. 

-----Original Message-----
From: rtg-bounces at lists.grdata.com [mailto:rtg-bounces at lists.grdata.com] On
Behalf Of Chris Miller
Sent: Monday, April 07, 2008 11:43 PM
To: John Von Essen
Cc: rtg at lists.grdata.com
Subject: Re: [rtg] Traffic is wrapping

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I've found if you optimize the database and run REPAIR TABLE and  
OPTIMIZE TABLE on a regular basis, things are MUCH more smooth. By  
optimize I mean adding indexes, tweaking queries and the like. Also,  
the query cache makes things go zooom.

As for pgsql -- why? I mean I know some would prefer that, but is  
there really any reason it would be a better fit for this?

Thanks,

Chris Miller
ServerMotion
www.servermotion.com

On Apr 7, 2008, at 10:19 PM, John Von Essen wrote:
> yrtg looks interesting... didn't know it existed.
>
> Does anyone know if anyone has tried to tackle a pgsql implementation
> with any success?
>
> Out of curiosity, at what point (say, interface count) does  
> scalability
> start to become an issue? Right now I monitor 30 FE interfaces  
> across 2
> routers (7600 series) with 5 minute polling - and haven't noticed any
> problems. Will I start to see some soon or is scalability problem much
> higher in the interface/device count range?
>
> Thanks
> John
>
>
>
> On Apr 7, 2008, at 8:38 PM, Ryan DiRocco wrote:
>
>> Also in regards to Bill's Yrtg there is a great 'yrtg-builer' that  
>> was
>> posted a few months back that actually works well.
>>
>> I honestly think .9 should have been grown out of .8-yrtg
>>
>> =======================
>> |  Ryan DiRocco
>> |  Director Of Engineering
>> |  Global Net Access, LLC.
>> |  PH:  404-230-9150 ext 240
>> |  Cell: 678-525-7307
>> |  E-Mail: ryan.dirocco at gnax.net
>> =======================
>>
>> -----Original Message-----
>> From: rtg-bounces at lists.grdata.com
>> [mailto:rtg-bounces at lists.grdata.com] On
>> Behalf Of bill fumerola
>> Sent: Monday, April 07, 2008 7:46 PM
>> To: Leech, Jonathan
>> Cc: rtg at lists.grdata.com
>> Subject: Re: [rtg] Traffic is wrapping
>>
>> On Tue, Apr 01, 2008 at 03:44:46PM -0600, Leech, Jonathan wrote:
>>> I think now is as good a time as any to further pollute the  
>>> waters...
>>> My
>>> colleague and I implemented JRTG to address the problems we had with
>>> RTG, primarily scalability. As its Java another benefit is there  
>>> is no
>>> problem building it.
>>>
>>> http://jrtg.sourceforge.net
>>>
>>> We built it to read the targets file from TargetMaker, and write to
>>> the
>>> same schema as RTG, but we never tested the database insertion as we
>>> write to a different schema.
>>
>> first: neat idea. i'm sure it works great better etc., no disrespect.
>> but..
>>
>> my lord, that code is so far gone from the original codebase... it's
>> nice that you gave a nod to its roots. i mean, i guess it's obvious
>> from
>> some of the structure and variables and event loops, but here is  
>> one of
>> many reasons why i stay away from java....
>>
>> %%%%%%%%%%%%%%%%%%%%%%%%%%%%
>> // need to make a new delegate if the className or delegateName for i
>> has
>> changed,
>> // or if there wasn't a delegate for i to begin with
>> if (
>> 	((oldDelegate != null) &&
>> (!oldDelegate.getClass().getName().equals(className) ||
>> !oldDelegate.name.equals(delegateName)))
>> 	|| ((oldDelegate == null) && (className != null) &&
>> !"".equals(className) && (delegateName != null) &&
>> !"".equals(delegateName))
>> 	)
>> {
>> %%%%%%%%%%%%%%%%%%%%%%%%%%%%
>>
>> OOP will save us all / save us all from OOP.
>>
>> if you had to rewrite it in an entirely other language, ruby would  
>> have
>> made a nice fit. dunno how advanced the SNMP modules are, but from a
>> language fitting the task perspective, i think it would work great.
>>
>> rrtg? bfrtg? odnsrtg?
>>
>> dare to dream.
>>
>> -- bill
>>
>> p.s. to the other poster who asked "Yahoo-RTG?"
>>      http://mu.org/~billf/yrtg/
>> _______________________________________________
>> RTG mailing list
>> RTG at lists.grdata.com
>> http://lists.grdata.com/mailman/listinfo/rtg
>>
>> _______________________________________________
>> RTG mailing list
>> RTG at lists.grdata.com
>> http://lists.grdata.com/mailman/listinfo/rtg
>>
>
> _______________________________________________
> RTG mailing list
> RTG at lists.grdata.com
> http://lists.grdata.com/mailman/listinfo/rtg

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (Darwin)

iD8DBQFH+undxBwlCB7CRwsRArHnAJ4p/x+UBwEZBcd3f5wrm2JZCTxtyACfVFOc
akjkriYOA0ttZnSChuW6RHQ=
=TdKI
-----END PGP SIGNATURE-----
_______________________________________________
RTG mailing list
RTG at lists.grdata.com
http://lists.grdata.com/mailman/listinfo/rtg