LISTSERV mailing list manager LISTSERV 16.0

Help for TECH-DISCUSS Archives


TECH-DISCUSS Archives

TECH-DISCUSS Archives


TECH-DISCUSS@LISTS.IPHOUSE.NET


View:

Message:

[

First

|

Previous

|

Next

|

Last

]

By Topic:

[

First

|

Previous

|

Next

|

Last

]

By Author:

[

First

|

Previous

|

Next

|

Last

]

Font:

Proportional Font

LISTSERV Archives

LISTSERV Archives

TECH-DISCUSS Home

TECH-DISCUSS Home

TECH-DISCUSS  November 2010

TECH-DISCUSS November 2010

Subject:

Re: pokey poke!

From:

Mike Horwath <[log in to unmask]>

Reply-To:

Technical Discussions List <[log in to unmask]>

Date:

Sat, 6 Nov 2010 14:59:08 -0500

Content-Type:

text/plain

Parts/Attachments:

Parts/Attachments

text/plain (43 lines)

And then I forgot about the biggest thing that I have beat up in the
last few months.

Monitoring and measurement systems!

I have decided that Zenoss, even with V3, is still way too much work
if you aren't running one of the L systems.  Way too many assumptions
about SNMP OIDs and the like.

New monitoring system is running in test mode now using Nagios v3 and
I implemented NagiosQL to help with day to day updates (no, we don't
do daily updates) so that others can help with the overall management.
This is working great and production testing will start very soon.  I
need to work on more NRPE stuff to threshold better for items running
on the servers themselves.  Windows monitoring sucks overall, though.

We still have Cacti up and running for our customers for doing
measurement, but threshholding is ick, but it has some great features
for per-user permissions and such.  Again, Windows monitoring is
pretty sucky.

Nick and I have been futzing hard with Zabbix as well (for
measurement).  Much potential for an agent based system as the agents
are small and unobtrusive.  SNMP is ... right out unless you wish to
build up a ton of templates for your *never* changing routers and
switches - they just aren't dynamic enough.  Thresholding is still
gonna require some more work but the on the fly graphing of <whatever>
you can grab with the agent is quite handy.  Unfortunately,
notifications when thresholds are broken is an instant notification so
transient 'issues' create quite a bit of noise still.  (we are working
on it) The really nice thing with Zabbix, because of the agent model,
the Windows systems can now be measured as well, including things that
were tough before, like disk space on that elusive L: drive.

(and OpenNMS was awesome but it is too much influx, requires way too
much set up - kind of like using RT as a ticket system, out of the box
it does something, but if you want it to work, be prepared for a lot
of time and sweat)

-- 
Mike Horwath      ipHouse - Welcome home!       [log in to unmask]
        The universe is an island, surrounded by whatever it is
        that surrounds universes. - Berkeley Fortune

Top of Message | Previous Page | Permalink

Advanced Options


Options

Log In

Log In

Get Password

Get Password


Search Archives

Search Archives


Subscribe or Unsubscribe

Subscribe or Unsubscribe


Archives

June 2011
May 2011
November 2010
April 2010
March 2010
February 2010
January 2010
December 2009
November 2009

ATOM RSS1 RSS2



LISTS.IPHOUSE.NET

CataList Email List Search Powered by the LISTSERV Email List Manager