Thanks. I looked more online as well and now getting the hang of what goes
behind it. My plan now is to atleast keep RRD files with full resolution till
the "last week". However, I am not able to still retain the max values across
the week. I think it is because the graph created for "one day" and "one week"
do not have their width as long as I would want them to show the full
resolution.
I am using the following reolutions in gmetad.conf . Could some one please tell
me where is the code in which I need to make the change in the graph width?
Thanks.
RRAs "RRA:AVERAGE:0.5:1:240" \
"RRA:MAX:0.5:1:240" \
"RRA:AVERAGE:0.5:1:5760" \
"RRA:MAX:0.5:1:5760" \
"RRA:AVERAGE:0.5:1:40320" \
"RRA:MAX:0.5:1:40320" \
"RRA:AVERAGE:0.5:720:240" \
"RRA:MAX:0.5:720:240" \
"RRA:AVERAGE:0.5:5760:365" \
"RRA:MAX:0.5:5760:365"
P.S. Assuming default 15sec intervals
Thanks,
Neel
On Mon, 28 Mar 2011 21:12:16 +0530 "Witham, Timothy D"
<timothy.d.wit...@intel.com> wrote
> This is normal, expected, and not a bug. Ganglia
consolidates data only by AVG, or average. MIN and MAX are not calculated or
stored. So when consolidating using MIN or MAX on a graph, you get the MIN or
MAX of the AVG. Data consolidation functions are decided at the time the
RRD files are created. You can’t add them in later. Adding MIN and MAX to all
the metrics would TRIPLE the size of the database. From: Indranil
Chakravorty [mailto:indran...@rediff-inc.com]
> Sent: Sunday, March 27, 2011 10:55 PM
> To: ganglia-general
> Subject: Re: [Ganglia-general] Wrong Max and Min values for Ganglia beyond
last hour time range... CONTD
>
> Thanks,
> Neel
>
> On Mon, 28 Mar 2011 09:24:52 +0530 "Indranil
Chakravorty"<indran...@rediff-inc.com> wrote
> >Adding one image at a time to keep the size of the mail within limit.
Please note the two images and see how the last year graph scales down the peak
value and in effect the entire historical data.
> >
> >Thanks,
> >Neel
> >
> >
> > On Fri, 25 Mar 2011 18:24:53 +0530 "Indranil
Chakravorty"<indran...@rediff-inc.com> wrote
> > >The problem ins persistent in the default graphs of Ganglia. For
example check these graph for "Network" for the last "day" and last "year" in
the attached images. You will notice that in the "last day" maximum there was a
peak of about 70 Mb/ps while the graph for the "last year" does not even scale
beyond less than 10 Mb and the peak indicated is about 8 Mb/ps. Ofcourse, the
MAX value should not reduce over the year. Anyone who could explain the
observation? Thanks a lot.
> > >
> > >Thanks,
> > >Neel
> > >
> > > On Fri, 25 Mar 2011 15:28:38 +0530 "Indranil
Chakravorty"<indran...@rediff-inc.com> wrote
> > > >Hi, I am plotting metrics by using gmetad. I am also
displaying the MAX, MIN, AVERAGE and LAST values on the ganglia graph. I
realised it today, that the MAX value is correct if you watch the graph for the
last hour. However, It is incorrect for any other time range. For eg. If in my
last hour the MAX value is 2900, however, when see the last day, week or year
the MAX value can be shown as lesser also. It seems ganglia simply gives the
average of the MAX values over time ranges beyond last hour. Could anyone
please provide me a workaround or a fix? I would be really grateful.
> > > >
> > > >Thanks,
> > > >Neel
------------------------------------------------------------------------------
> Create and publish websites with WebMatrix
> Use the most popular FREE web apps or write code yourself;
> WebMatrix provides all the features you need to develop and publish
> your website. http://p.sf.net/sfu/ms-webmatrix-sf
> _______________________________________________
> Ganglia-general mailing list
> Ganglia-general@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/ganglia-general
>
------------------------------------------------------------------------------
Enable your software for Intel(R) Active Management Technology to meet the
growing manageability and security demands of your customers. Businesses
are taking advantage of Intel(R) vPro (TM) technology - will your software
be a part of the solution? Download the Intel(R) Manageability Checker
today! http://p.sf.net/sfu/intel-dev2devmar
_______________________________________________
Ganglia-general mailing list
Ganglia-general@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ganglia-general
