Maximillian Dornseif wrote:
I administer about a dozen FreeBSD Servers. This results in me
getting about 100 mails a week from the PERIODIC(8) scripts.
Obviously this is to much to read with care.
I wonder what the canonical approach is to handling hundreds of
status mails like the ons generated by periodic.
Yeah. Periodic scripts are great if you only have a few machines not
doing anything really critical. It's a mark of the standards required
by the FreeBSD project that the baseline system comes with a working
monitoring system built-in.
However, periodic e-mails really do not scale to tens, let alone hundreds,
of machines and once a day is really far too infrequent for checking
important services.
My recommendation is to redirect all of the periodic output to local log
files -- daily.log, weekly.log etc. -- according to the comments in
/etc/defaults/periodic.conf. Then set up a full blown monitoring system
using eg. Nagios. For a dozen or so servers, you won't need anything
particularly special to run Nagios on. Sticking it in a jail on one of
your existing machines might work well for you.
Nagios monitoring needs to be developed over time. With the standard
plugins supplied by the net-mgmt/nagios-plugins port, you'll be able
to monitor most services with external listeners. However, to get the
most out of Nagios I find that installing net-mgmt/net-snmpd on all
monitored machines is a necessity. You'll need to be careful about
how you do that -- SNMP v2c or lower does the equivalent of sending
passwords across the net in plain text, so it's useful if you can
arrange for some sort of private back-end network between all your
servers.
While the intrinsic capabilities of net-snmpd add a great deal of
monitorable information, the real advantage is the ability to hook
up arbitrary scripts via the 'extend' mechanism. This coupled with
the ability to write custom plugins for Nagios means you can do a lot
of very interesting things. Trivial example: we routinely hook up
running 'gmirror status -s' through snmpd and then use a very short
Nagios plugin script to alert us to RAID problems.
There's also nagios-exchange with plenty of interesting and useful stuff
available -- see for instance
http://www.nagiosexchange.org/cgi-bin/page.cgi?g=Detailed%2F1562.html;d=1
Nagios and periodic don't fulfil exactly the same functions, but there
is enough overlap to allow replacing one with the other.
Cheers,
Matthew
--
Dr Matthew J Seaman MA, D.Phil. 7 Priory Courtyard
Flat 3
PGP: http://www.infracaninophile.co.uk/pgpkey Ramsgate
Kent, CT11 9PW
signature.asc
Description: OpenPGP digital signature
