2008/4/30 richard lucassen :
> > We're just starting the process of rolling out the next (?last)
> > release in the 5.2.x line. It might be worth grabbing the
> > 5.2.5.pre2 tarball from the project website, and seeing whether
> > that also shows the same problem.
> > Similarly for the most recent release (5.4.1).

>
> Well, I think the whole problem has nothing to do with snmpd. The
> problem was a sundance driver problem. The problem just shows up when
> running snmpd ;-)



But is that (at least in part) because snmpd is playing nasty?

Checking the relevant code, it seems to be lifted verbatim from
the "mii-diag" tool, which dates from around 2003. Even then,
the original code talks about "new or old SIOCGMIIPHY" values.
(with SIOCDEVPRIVATE being the "old" value).

Five years on, I'm tempted to suggest that there is no point in
supporting this (presumably) transitional/experimental old ioctl.
It would be safer (as well as simpler) to drop mention of
SIOCDEVPRIVATE altogether, and just use the SIOCGMII*
ioctly (and refer to them by name, rather than number!)


I'll float this idea on the -coders list, and see what the other
developers think.

Dave

-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
Don't miss this year's exciting event. There's still time to save $100.
Use priority code J8TL2D2.
http://ad.doubleclick.net/clk;198757...un.com/javaone
_______________________________________________
Net-snmp-users mailing list
Net-snmp-users@lists.sourceforge.net
Please see the following page to unsubscribe or change other options:
https://lists.sourceforge.net/lists/...net-snmp-users