Hi There,
As you might have noticed from the nagios-users list, I have been trying
to bend nagios into doing what I want, with limited success. It is
probably something that nagios was not intended to do, so I mean no
insult to this fine project by so saying.
I wanted to check a couple of things with you before carrying on.
1 - Are you guys open, in principle, to me doing some work on the source
tree. If the answer is that it is very complex cod and that you would
prefer it if the core developers were the only ones doing the work then
I would quite understand. If the answer is no, then please take the
following as RFEs rather than requests for permission for me to do the work.
2 - The message I sent yesterday outlined a confusing name change for
the service variable active_checks_enabled to checks_enabled. I would
like to go through the code and replace instances of the relevant
checks_enabled name to active_checks_enabled. I quite agree that once
you have found the link between these variables you can easily treat
them as being the same thing at the service level. However I would
suggest that other people new to the code would fall into the same
misunderstanding that I did. It lost me a deal of time in trying to work
out why things were doing what they were doing.
3 - Are you guys interested in moving to the automake facility. You are
currently using autoconf to create the Makefiles (and others). This
requires the maintenance of a set of quite large and intricate
Makefile.in files. With automake there would be a set of far smaller and
simpler files to maintain .. plus you would gain the developments that
come for free with automake. An example of an advantage would be the
addition of the "make dist" facility, which, in one hit, creates a
tar.[gz|bz2] file. Given that you already support a .spec file, it would
allow a very easy transition from source tree to rpm fileset
"make dist; rpmbuild -ta "
4 - The current CVS willnot build clean on my RH80 build machine. It
wrongly adds -lcma to the link line. This would seem to be an HP/UX only
library and this problem was not present in the 1.0 release. When I hand
edit the final Makefile and remove this flag, the system builds clean.
Is this a known and in progress problem, or would you like me to
investigate it as I use Linux and not HP/UX.
5 - This is more Blue-Sky, but it would be very interesting if as well
as nagios servers taking check results from plugins and remote nagios
servers, they could also take fill blown state change lines (as per
status.log) from another nagios server. This would allow a set of sites
to effectively mirror the state of other sites in the group. I
particularly want this (or equivalent) as I look after a number of
customer sites. I would like each of these sites to have their own
stand-alone nagios setups, but to have my nagio server act as a
composite of them all. I don't want them to be able to see each other's
states for obvious reasons. I realise that I could piggy-back the obsess
mechanism to mirror check results, but in the event of a network outage
between their site and mine, I would lose valuable state change history
for the at least part of the outage.
I would be very grateful if you folks could have a think about the above
items and let me know how you feel.
Cheers
Andrew Meredith BEng CEng MBCS MIEE
_______________________________________________________________
The Anvil Organisation Ltd.
Director
Tel: +44 (0) 1249 444240 | Email: andrew@anvil.org
Fax: +44 (0) 1249 460560 | Web: http://www.anvil.org/
Mob: +44 (0) 7802 389007 | WAPMail: andrew.meredith@orange.net
_______________________________________________________________
The box says Win95 or better .. Must run under Linux then!
This post was automatically imported from historical nagios-devel mailing list archives
Original poster: andrew@anvil.org