~/hobbit/client/* is where the client stuff lives. The server runs a
local client. On remote systems, you usually clone the ~/hobbit/client/*
part to other machines (assuming the same architecture). As you get more
into it, you will figure out how to make client distributions.
If you are seeing anything reported for CPU, memory, disk, etc. on your
server, then the client is running. You don't have to do anything to enable
it.
*From:* Scott Mohnkern [mailto:user-1386a26b8c7d@xymon.invalid]
*Sent:* Wednesday, October 31, 2007 3:52 PM
*To:* user-ae9b8668bcde@xymon.invalid
*Subject:* Re: [hobbit] New to Hobbit --- file monitoring
This may be where I'm getting confused. On the machine where hobbit is
reporting data, (i.e. the machinename I use for the url) does the client
need to be running? Where are the config files for the client typically
stored? I've been editing files /etc/hobbit But I wonder if those are just
the server files, and not the client files (assuming the client needs to be
running as well on the machine that is the server.
Scott
On 10/31/07, Stef Coene <user-dbffe946c0f4@xymon.invalid> wrote:
On Wednesday 31 October 2007, Hubbard, Greg L wrote:
How long did you wait? It takes a bit (10 minutes or more) for
changes
in client-local.cfg to propagate to the clients (they have to detect
the
change, and then pull it in, then act on it, and this takes a couple
of
poll cycles).
I tried to set up file monitoring today and after 1hr, I fixed it by
restarting the hobbit server. Not reloading, restarting. So, restart
your
hobbit server after you changed the config files and the change is not
detected.
You can also take a look in the tmp directory of the client. There
should be a
logfetch file if the client picks up the new configuration. There
should be
2 files, one with the settings on the hobbit server and one with the
status
(this is a file used by the hobbit client to track the logfile).
Stef