[Orca-users] (no subject)
sean at seanoneill.info
Tue Dec 10 08:19:01 PST 2002
At 09:12 PM 12/9/2002 -0600, Liston Bias wrote:
>On Mon, 9 Dec 2002, Sean O'Neill wrote:
> > At 02:04 PM 12/9/2002 -0800, Mark Goldenberg wrote:
> > >Has anyone used 'scp' to copy the orcallator.se generated files to the
> > >orca server? I don't use NFS in my environment and I'm looking for
> > >another was to move the files?
> > >
> > >If I could use 'scp' could I add the code to the orcallator.se file?
> > As Blair said, you probably could modify the COMPRESS variable to do
> > this. But you lose the "almost realtime" aspect of Orca doing it this way.
> > I use the following script on my Orca server to PULL the data from my
> > remote servers. You notice I said PULL and not PUSH. I let the remote
> > server simply collect the data - it shouldn't be their responsibility to
> > PUSH data to the Orca server. What's happens if the Orca server is down
> > ? Using a PULL configuration, this greatly simplifies the data transfer
> > requirements for error detection - IMHO.
>We actually use PUSH. It seems to be much better on the performance side
>and I didn't want orca having to keep track of systems that happen to up.
Curious what you mean here by better performance using a PUSH model vs a
PULL model. One strong distinction is the data "gets there faster".
Another reason I did a PULL config is I have a Ultra5 acting as the Orca
server. Its already pegged at 100% utilization 100% of the time. By using
a PULL, I limit the impact of data transfer on CPU utilization. I would
rather the CPU be crunching RRD and graphs then 50 rsyncs.
>We are regularly adding servers and are currently graphing about 250
>systems on one server with 5-10 minute delay in graphs. I actually
>install them though jumpstart:
Kewl, thanks for sharing the wealth :)
......... ..- -. .. -..- .-. ..- .-.. . ... ............
.-- .. -. -... .-.. --- .-- ... -.. .-. --- --- .-.. ...
More information about the Orca-users