[rtg] RTG Bug Fixes

[Bryan Wann]

On Wed, 8 Oct 2008, Jeff Bacon wrote:

> Not really. I have a feeling there are a lot of us who are sorta just
> poking at it here and there and keeping it going for what we need/use it
> for, but that's about it. And not everything seems to work for everyone;
> for instance, I'm still using 0.7.4 rtgplot against the 0.9 poller/db
> because the 0.9 rtgplot goes into la-la-land when compiled and run on
> sol10/sparc/gcc[234]. I'm sure I could figure it out but it's just not
> worth the time and effort.
>
> It would be interesting to aggregate all of the various mods/changes we
> have individually done just to see what is out there, though.

I pretty much gave up on trying to keep RTG running. For the past 4-5 
months I've been using the JRTG (http://jrtg.sourceforge.net/) poller. As 
previously discussed, the code is pretty far removed from rtgpoll but just 
similar enough to make it an easy drop in replacement.  Apparently I'm the 
only one publically admitting I use it in production. It works beautifully 
keeping up with 122,000+ targets under 80 seconds, and the majority of 
those are slow MSFC-less sup1/sup2 6513s scattered all over the globe.

I use btoneill's targetmaker every day.  There seems to be a couple of 
bugs, like it forgets targets after using --update and no snmpbulkwalk?! 
I haven't had a chance to sit down and make it better yet, but it's 
quickly getting to the top of my list.  Once I figure out a replacement to 
rtgplot, there won't be much of the original RTG distribution left that I 
actually use.


Met vriendelijke groet/kind regards,
bryan