[Xymon] Modernizing the DNS check

[Mark Felder]

On Tue, Jan 7, 2014, at 23:07, Jeremy Laidman wrote:
> Yes, lots of good rebuttals there.  I think I have to agree that most of
> my
> proposed checks are quite niche, but the one you've proposed is probably
> the least niche of the lot.  So if a sizeable number of Xymonsters could
> make of it without needing too much configuration (hence complex parsing
> code), then it's worthy of inclusion.
> 
> On 8 January 2014 13:44, Mark Felder <feld at feld.me> wrote:
> 
> > The question here is "Are the publicly accessible NS servers in a
> > consistent functional state?". The goal is not to validate the data.
> 
> 
> So, I suppose the "object" you're trying to watch is the
> "NS" consistency state of the zone.  So yes, you'd alert against the zone
> name such as what you've shown in your hosts.cfg example.
> 
> Although I can't speak for Henrik's design model, I do think that
> xymonnet
> is not geared up to monitor this type of object, and instead it expects
> its
> objects to all have either one IP address, or a name that resolves to an
> IP
> address.
> 
> So really, I think the answer to your original question is that the DNS
> check capability probably can not be easily enhanced to check something
> that doesn't look like a host.
> 
> But there's no reason I can think of that some zone check code couldn't
> be
> added into some other part of Xymon.  But probably not as an extension to
> the existing DNS check.
> 
> Many of the internal checks seem to have been modelled on ext scripts.
> 
> J

Thanks for the feedback on everything. It's nice to know I'm not talking
to myself on this list :-) You've also made me consider a few scenarios
I previously didn't carefully consider, but I have some other ideas on
how to prevent those errors by enhancing the procedure used to change
DNS entries.