I wouldn't hold my breath to see any form of notification pop up in 2.x
(unless, of course, someone's about to spring something on everyone).
3.x is being designed with a more open framework and, although it's still
fairly early along to tell for sure, should at the very least be able to
support monitoring and nicer metric features via plug-ins. There aren't
too many plug-ins right now since the API's still gelling. :)
Please don't forget that gmetric is a very simple program that reads your
local gmond.conf and command-line arguments, makes an XDR packet, and then
sends it on the multicast channel according to those specs. If you want a
metric to be recorded over time, a cron job or scheduler of some sort is
necessary. gmetric does not communicate with the local gmond directly in
any way (the local monitoring core "finds out" about the metric through the
multicast channel in the same way remote ones do).
I think the original poster's best bet in this case (to get notification)
is to hack up that perl script to do it...
An alternative, which some people seem to be using to great effect, is to
have Nagios do all your notification and Ganglia do all your data
collection. It's not my personal setup so I can't comment on it...
Steven A. DuChene wrote:
ganglia does have any means of re-polling/re-reading a metric. I have several
extra metrics I have added to my cluster and I have cron jobs that periodically
resubmit the metric. How often that occurs depends on what the metric is and
how volatile it is. Some I refresh only every 30 minutes and others more often.
Check the list archives for the alarm thing. There was a patch posted a while
ago to enable this sort of thing but i don't believe it has ever been mentioned
again so I don't know if it is ever going to be integrated into the main
ganglia source tree.
Perhaps Matt or one of the other developers could answer that.
-------Original Message-------
From: [EMAIL PROTECTED]
Sent: 02/14/03 11:27 AM
To: [email protected]
Subject: [Ganglia-general] Problem with added metrics
Hi,
All my ganglia works perfectly except the metrics I added, like cpu
temperature. Ganglia reads the metrics one time never more, except if I restart
the metrics.
I use a perl script to get the metrics from motherboard monitor and gmetric to
add it to ganglia.
Also I want to know if it is possible to set some alarms with ganglia.
Exemple : If the cpu temperature of one node goes over 72 celcius.
Tanks
Karl
-------------------------------------------------------
This SF.NET email is sponsored by: FREE SSL Guide from Thawte
are you planning your Web Server Security? Click here to get a FREE
Thawte SSL guide and find the answers to all your SSL security issues.
http://ads.sourceforge.net/cgi-bin/redirect.pl?thaw0026en
_______________________________________________
Ganglia-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ganglia-general
-------------------------------------------------------
This SF.NET email is sponsored by: FREE SSL Guide from Thawte
are you planning your Web Server Security? Click here to get a FREE
Thawte SSL guide and find the answers to all your SSL security issues.
http://ads.sourceforge.net/cgi-bin/redirect.pl?thaw0026en
_______________________________________________
Ganglia-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ganglia-general
