[Orca-users] Re: Format of input file for Orca

[Blair Zajac]

Einar Jóhannesson wrote:
> 
> Thanks Blair.
> I had already reduced my data to a two column file "timestamp
> InOctets".
> 
> I was not shure orca would read just one line or any number of data
> lines.
> I then used the orcallator.cfg under $prefix/lib as a sample
> plot {
> title                   %g Interface Bits Per Second: $1


If you use the $1 format, then you need to use the () construct around
the particular data you are plotting, something like

    data 8 * (InOctets)

The %g is a lower case version of the source group name.  This comes
from matching portion of the filename.  So if you had

    find_files /input/data/(.*)/input.txt

and this matched a file

    /input/data/host12/input.txt

then %g would be host12

> source                  orcallator
> data                    8 * InOctets
> line_type               area
> legend                  Input
> y_legend                Bits/s
> color                   00ff00
> data_min                0
> data_max                10000000
> 
> I put my input file into
> $prefix/var/orca/orcallator/percol-yyyy-mm-dd.
> I then tested orca and .......it worked.
> I don't know any perl yet, it seems I will soon.
> 
> I now have to understand how the %g and $1 are used in the graphs.
> 
> Then I will need to plan how to split up my data for each host,
> snmpCollected instance and mib variable
> (I have all data for ifInOctets for interface 1 for all hosts in one
> file called ifInOctets.1 and then ifInOctets.2 for interface
> no. 2)
> maybe use
> $prefix/var/orca/orcallator/host/MIBvariable.instance-yyyy-mm-dd
> as the output format when I dump the snmpCollected data.

Yes, this is what you want to do.  Then use a find_files like

find_files
$prefix/var/orca/orcallator/(.*)/MIBvariable.instance-\d{4}-\d{2}-\d{2}

You'll have to replace $prefix with the real prefix you used, since
there's no substitution in this file.

Blair

> 
> Einar
> 
> blair at gps.caltech.edu wrote:
> 
> > Hi Einar,
> >
> > You should be able to read this data in pretty easily.  Use the
> > orcallator/orcallator.cfg.in file as a basis to start writing your
> > own Orca configuration file for your data.
> >
> > The one thing you will need to change is the lines
> >
> >     column_description      first_line
> >     date_source             column_name timestamp
> >
> > to something like
> >
> >     column_description      date time hostname count epoch
> >     date_source             epoch
> >
> > and then for a plot of this type
> >
> > plot {