MIB-RFC1213 is already in the default smi_modules file... I am aware
that asking libSMI to reload a MIB already loaded (even if renamed)
causes a crash (I see oids.c:572 in the stack but that's a debugger
issue I guess). But I do not know a simple solution to avoid
reloading a MIB that is already loaded, I've open a bug with the
I have an item in http://wiki.wireshark.org/Development/SNMP about the issue...
This weekend I'll try to close as many of the critical Items in that list.
On 11/8/07, Andrew Hood <[email protected]> wrote:
> Ulf Lamping wrote:
> > Gerald Combs schrieb:
> >>It's been a while since 0.99.6 was released (Holy crap! July 5!), so it would
> >>probably be a good idea to push 0.99.7 out the door. I'd like to make a
> >>prerelease on the 19th or 20th, with a final release a week later.
> > I had the same idea recently.
> > Things that probably need a fix before:
> > - buildbot is unhappy (make it happy again somehow)
> > - SNMP changes finished?
> I build HEAD yesterday on Windows (MSVC2005).
> If I include RFC1213-MIB, required to import various Microsoft MIBs, it
> segfaults in oids.c at line 574.
> If I don't include RFC1213-MIB it generates an error about the MIB being
> missing, but at least it runs.
> If I run in under the MSVC debugger some of the time it works with
> RFC1213-MIB present, some time it doesn't.
> I've run smilint against all the M$ MIBs and fixed all the "errors",
> event the cosmetic ones, and it still segfaults.
> There's no point in being grown up if you can't be childish sometimes.
> -- Dr. Who
> Wireshark-dev mailing list
> Wiresh[email protected]
This information is top security. When you have read it, destroy yourself.
-- Marshall McLuhan