[Orca-users] Re: Orca plots on one central server

[John Mastin]

Hey there Mark!

Basically what we do here is run an orca client piece that I put
together.  On the clients, we install the SE Toolkit, rsync and the
orcallator.se file with the necessary start/stop scripts (we have
created Solaris packages of rsync and orcallator).  We publish the
percol data files via anonymous rsync (you can horse around with
authentication with rsync if you'd want to make sure hax0rs are not
looking at your performance data).  On the central orca server, we use
rsync to grab the files from the clients and then we munge it into rrd's
and html files.

Hope this helps!
Johnny

> My company doesn't want to renew their TeamQuest
> capacity planning software license, so I'm thinking
> about using Orca with orcallator.se.  However, I have
> one question about its setup:
>
> Do I have to NFS mount all the clients' orcallator
> files onto the Orca server to get the plots in one
> place?  Is there some standard method that you guys
> are using, so that you can have one central server
> where Orca will plot the graphs, and have all the
> clients running orcallator.se and sending the data
> over to the central server?
>
> Your help in this would be greatly appreciated.  I'd
> hate to have to setup some kind of cron job on each
> and every box to rdist the files over.  There's got
> to be a better way.  Thanks in advance.
>


--
John Mastin, Jr.              email: john.mastin at bms.com
Bristol-Myers Squibb            phone: (609) 818-3788
PRI Infomatics              fax: (609) 818-7693
Princeton  NJ