Concurrent transactions - ragraph question

Rafael Barbosa rrbarbosa at gmail.com
Thu Jun 17 10:29:57 EDT 2010


I have being discussing the results with some colleagues here. Probably the
result I am having is flows/sec, and not the total number of flows in the
period. I believe the program counts the number of active flows in the
interval (in my case 5min) and calculating "flows/sec".

I will run with the debug option to see if I can confirm this (or not)...

Thanks for the timely reply.

Rafael

On Thu, Jun 17, 2010 at 4:17 PM, <carter at qosient.com> wrote:

> Hey Rafael,
> That maybe an artifact of rrd and rrd_graph.
> You can run ragraph with --debug and it will show you all the numbers get
> into the rrd.
>
> The particular command you are using is the simple example and the least
> efficient. Try "-m srcid" and the numbers may get around the rrd issues, and
> the debug output should help you see what is gong on.
>
> Carter
>
> Sent from my Verizon Wireless BlackBerry
> ------------------------------
> *From: * Rafael Barbosa <rrbarbosa at gmail.com>
> *Date: *Thu, 17 Jun 2010 11:58:47 +0200
> *To: *<argus-info at lists.andrew.cmu.edu>
> *Subject: *[ARGUS] Concurrent transactions - ragraph question
>
> Hi all,
>
>
> I have been testing some of the argus examples present in the wiki, and one
> of the outputs called my attention. After analyzing the output of:
>
>
>  ragraph trans -M 10s -r $file - title 'Concurrent Transactions' -w transac.png2
>
> for a file I have, I get a graph with the number of concurrent
> transactions, but for my surprise I don't get an integer number of
> transactions. How is it possible to have around 13.2 concurrent
> transactions? How does argus calculate this value?
>
> By the way, there is a small typo in the the wiki for this command. There
> is a space in "- title".
>
>
> Best regards,
>
> Rafael Barbosa
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://pairlist1.pair.net/pipermail/argus/attachments/20100617/a11ce94f/attachment.html>


More information about the argus mailing list