[TriLUG] load monitoring, logging on rhel/centos

[Cristóbal Palmer]

I've used Nagios before and will continue to use it, but more than
monitoring I'm interested in logging. What are you using to log load
(OS load, application load, etc)?

I'm especially interested in solutions that are focussed on logging or
parsing of logs of mysql/apache on RHEL5. Summaries and graphs that
can be shared with the PHB would be terrific, but text-only solutions
may still be useful.

Here's the scenario in a bit more detail. I've got many websites
sharing several mysql servers, and am not always able to find time to
actively troll the slow query logs. I'd like something that gives me a
bit of advance warning about what sites are getting hit how hard and
how they're handling it. A site maxed out our mysql connections for a
three minute blip? Great... when exactly did that happen, which site
was it, etc. etc. Is there a project out there that simplifies
answering these kinds of questions? If I see one of my mysql servers
blip in Nagios, I don't want to have to go digging to figure out why
it did so. Even when there are no problems, logs of what load was
caused when by what site would be handy to have.

Cheers,
-- 
Cristóbal M. Palmer
"Small acts of humanity amid the chaos of inhumanity provide hope. But
small acts are insufficient."
    -- Paul Rusesabagina