Steven,
sorry to bug you with this comment, you are mentionend in the
HOST-RESOURCES-MIB as the responsible contact.
I hope i am addressing the right person for this subject, if not, it would be
nice if you could forward this email to the right WG, since the "IETF Host
Resources MIB Working Group", which is mentioned in the MIB, does not seem to
exist any more.
Maybe my whole request/complaint is out of date or even redundant, but i
wanted to give my feedback anyway.
There is some aspect of the HOST-RESOURCES-MIB really chewing on me.
I am just extending the NET-SNMP daemon, especially the hrSWInstalled tree of
the HOST-RESOURCES-MIB, for a specific Linux-Distribution which is not
supported yet by this daemon.
However, i stumbled over the "uselesness" of the hrSWInstalledID-Object. It is
used to give out the version of the installed software-package.
My problem is that it allways implies that the distributor maintains his
Enterprise-MIB-space and extends it every time a new software-package is
included.
As you can imagine, this doesn't make sense in most cases - with no UNIX, no
Windows, no Linux, no nothing. Well, maybe Cisco likes to do that, but in
general it is very unpractical since nobody is going to replace his MIBs just
because "Foobar Inc." in Southwest-Kickass, Greenland, released version
3.1.4.22-rc4-beta of their software-package "Fridgecontrol".
Every vendor seems to set it to SNMPv2-SMI::zeroDotZero - i don't know about
their motivation, but i think they have the same idea like me.
I wanted to know if there are any plans to *replace* the variable in the
hrSWInstalledID-Object (currently ProductID which points to the
Enterprise-space of a vendor) with something useful, like an IA5 String? I
could easily set this string to the current version of the software-package if
it would be IA5 or something else.
I know, i am just one in a million SNMP-users, but it would be really nice if
you could bring up this topic if a 2790bis is being considered. I know that
someone just can't change a MIB - there is too much depending on it - but the
dependencies should be *really* in the minority this time.
Thanks a lot for your time,
sincerely, Alexander Janssen.