On Wed, 2005-11-02 at 08:01 -0800, Wes Hardaker wrote:
> I think that 5.2.2 is looking less and less useful to the
> end user which makes me more and more wary of releasing it.

> But given that embedded perl is broken currently and because
> snmpnetstat would have to be non-existent in it, I'm thinking
> that we shouldn't release 5.2.2 as is.

I'm very sympathetic to that.
In which case, I think we should stand down from "release
candidate" status, and revert to pre-release mode.   Whether
that means 5.2.2.pre4 or 5.2.2.1.pre1, I don't really care.



>                               We should fix both of these problems
> (it sounds like Dave is really close, and we have a patch for perl)

I'd have no objection to reverting to preN, and fixing the perl
problem.  I'm not so sure that my re-write of snmpnetstat would
be appropriate for 5.2.2.   Remember that this is basically a
bug-fix release - the user assumption will be that is a drop-in
replacement for previous 5.2.x versions.

That's not true of the new snmpnetstat utility.  Because I'm now
using the standard 'snmp_parse_args' routine, several of the
previous command-line options are no longer available (or do
something different).
  I don't particularly like breaking backward compatability like
this, but it does have the advantage of emphasising that this is
a "new" (legal) utility, replacing the previous (non-legal) one,
as well as dramatically improving the functionality of the thing.

Changing (and possibly breaking) something as part of a new
major-release seems reasonable. Not ideal, but understandable.
Changing/Breaking it as part of a bug-fix release is much less
justifiable, IMO.   Hence my vote ranking earlier today (and
the position of my "veto" vote).

   If we simply drop this tool from the 5.2.x line, then anyone
with an existing 5.2.1 installation can continue using the previous
version, with exactly the same behaviour as before.  If we include
the new re-write, then installing it will break any existing usage.

It's not ideal - but I don't think there is a "good" option open
to us.  This feels to be the least-worst option, IMO.

Dave


-------------------------------------------------------
SF.Net email is sponsored by:
Tame your development challenges with Apache's Geronimo App Server. Download
it for free - -and be entered to win a 42" plasma tv or your very own
Sony(tm)PSP.  Click here to play: http://sourceforge.net/geronimo.php
_______________________________________________
Net-snmp-coders mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders

Reply via email to