Re: [Nagios-devel] Custom Object Variables: Contrary to docs,

Support forum for Nagios Core, Nagios Plugins, NCPA, NRPE, NSCA, NDOUtils and more. Engage with the community of users including those using the open source solutions.
Locked
Guest

Re: [Nagios-devel] Custom Object Variables: Contrary to docs,

Post by Guest »

On 2010-08-28 15:59, Ethan Galstad egalstad-at-nagios.org |Lists/Send to=20
lists| wrote:
> Thanks for pointing this discrepancy out. Custom macros should be
> cleaned IMO, so I opted to update the docs instead. :-)

During the weekend I haven't been able to understand the rationale=20
behind this. Could you help me understand it?

Looking at the other macros subjected to macro cleansing[1]

1. $HOSTOUTPUT$
2. $LONGHOSTOUTPUT$
3. $HOSTPERFDATA$
4. $HOSTACKAUTHOR$
5. $HOSTACKCOMMENT$
6. $SERVICEOUTPUT$
7. $LONGSERVICEOUTPUT$
8. $SERVICEPERFDATA$
9. $SERVICEACKAUTHOR$
10. $SERVICEACKCOMMENT$

it looks to me as if the values of these all originate from outside the=20
config files. Then cleansing makes sense: "We don't really know/trust=20
the source of these values, so lets make sure they are safe".

But the values of Custom Object Variables come from the config files, so=20
why aren't they to be trusted? I don't (yet) see the conceptual=20
difference between allowing special/illegal characters in Custom Object=20
Variables and allowing them in $ARGn$ definitions. If we don't trust the=20
author of the config files, shouldn't we cleanse $ARGn$ definitions in=20
"check_command"s too then? Why one and not the other?

Peter
--=20
Peter Valdemar M=C3=B8rch
http://www.morch.com





This post was automatically imported from historical nagios-devel mailing list archives
Original poster: ?UTF-8?B?IlBldGVyIFZhbGRlbWFyIE3DuHJjaCAoTGlzdHMpIg==?
Locked