[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [hobbit] get hobbit to show status for same host but with different configuration
- To: hobbit (at) hswn.dk
- Subject: Re: [hobbit] get hobbit to show status for same host but with different configuration
- From: Henrik "Størner" <henrik (at) hswn.dk>
- Date: Mon, 12 Jan 2009 12:59:06 +0000 (UTC)
- Newsgroups: lists.hobbit
- Organization: Linux Users Inc.
- References: <gkf9ij$dru$1 (at) voodoo.hswn.dk> <1ebd03650901020113u39d2d030p37d8e4f9d550c6d5 (at) mail.gmail.com> <gjt06m$7u6$4 (at) voodoo.hswn.dk> <1ebd03650901050530g5cf3859awc5742295170e070c (at) mail.gmail.com> <gkf9ij$dru$1 (at) voodoo.hswn.dk> <1ebd03650901120405j3dc70937v7969cf353103cdcb (at) mail.gmail.com>
- User-agent: nn/6.7.3
In <1ebd03650901120405j3dc70937v7969cf353103cdcb (at) mail.gmail.com> "Mogens Busk" <mogens.busk (at) gmail.com> writes:
>I think that Xymon works this way:
>Xymon server receives client information from the xymon-client, stores this
>information and then generate the xymon webpage on request.
>When xymon server is asked for a specifik page for clienta+proc, the xymon
>server check what to do with the stored information against the
>server/etc/hobbit-client.cfg file and the generates the webpage.
Not correct. Xymon receives the client information from xymon-
client, sends this through the hobbitd_client module which reads
the configuration and generates the cpu-, disk-, procs- etc. status-
messages.
So once the client message has been processed, the status is fixed.
The fact that there's a CGI script which generates a webpage from
the status doesn't change that.
>basicly I want to make 1 page showing everything of interest for the
>sysadmins and another one only showing part of the clienta+proc specifik
>for a given customer.
>I have Proc, Ports and Services for many customers on the same server and
>would to be able to tell which customer is affected when some monitor goes
>red.
Unfortunately that is not possible right now. It is an enhancement I
would like to make that you could create multiple "procs" messages
with different groups of processes - that would solve your problem,
and would also allow us to focus alerts more directly to the application
team responsible for a specific application. Same method could be used
for e.g. "disk" status.
Regards,
Henrik