DS> I've just committed a new implementation of the Disman Event MIB DS> to the main CVS development branch. I'd like to propose that this DS> should replace the existing implementation for the upcoming 5.3 DS> release and above.
> This makes me a little nervous, as we had talked about (and I thought > decided) to turn on this mib by default. > Yes - that's what triggered this flurry of activity. But surely what we're interested in providing (by default) is the Event-MIB *functionality*, rather than a particular implementation? Wes said some time ago that he really like to rework this MIB using the v5 APIs. And describing the DisMan MIBs for TBB revealed a number of limitations and deficiencies in the current implementation. Rather than documenting these problems, I decided to re-implement the MIBs correctly (partly because I couldn't concentrate on the book!) And the decision to enable Event-MIB support by default raised the priority of doing this. > Might I suggest a configure flag (--enable-disman-rewrite) for 5.3, > with the switch to the new version as the default in 5.4? What's the point of that? If this new version is better, then why wait to bring it into service? If it's not better, then I'm wasting my time, and should just scrap the code completely. We've learned that the only way to improve an implementation is to bring it into service (as part of the default configuration) and deal with the bugs that surface. Anything "optional" tends to get ignored 99% of the time (as I discovered to my cost with the mibJJ fiasco). You must have found this with the IF-MIB rewrite! Given that the existing Event-MIB code isn't currently part of the default build (at least for any released versions), now seems the optimum time to switch to a replacement - *before* the old code starts to get more widespread. It's not quite the same situation as the IF-MIB rewrite, which involves very some very architecture-specific code. All the DisMan code (apart from RemOps) is inherently architecture-independent. (The values it works with may not be, but the basic operations are). So if it works on one system, it should work across the board. But there's only one way to find out for sure. Dave ------------------------------------------------------- This SF.Net email is sponsored by: Power Architecture Resource Center: Free content, downloads, discussions, and more. http://solutions.newsforge.com/ibmarch.tmpl _______________________________________________ Net-snmp-coders mailing list Net-snmp-coders@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
