Doc Schneider writes:
> Justin Mason wrote:
> > rc2 seems pretty solid. I suggest if this keeps up 'til tomorrow, we
> > should put it out as 3.2.0-beta1?
> >
> > The remaining bugs in the 3.2.0 queue are:
> >
> > 4834 nor P3 NEW Patch: SA 3.1.1 "make test" fails
> > in jail on FreeBSD 5.2.1-RELEASE
> > 4481 nor P5 NEW if ($@) { dbg constructs *should*
> > report the value of $@
> > 5420 nor P5 NEW DnsResolver.pm fails when new()
> > returns EACCES
> >
> > These 3 are trivial in scale. However I'd prefer not to apply them unless
> > we run into another serious bug which requires another rc tarball; any
> > change can cause problems at this stage ;) They could all wait for 3.2.1
> > in my opinion.
> >
> >
> > 5412 min P5 NEW deferable spamc -x -R always
> > returns zero
> > 5419 min P5 ASSI deferable kill -HUP `pidof
> > spamd` causes the ps name to change from "spamd" to "perl"
> >
> > These we can safely leave for 3.2.1, I think.
> >
> > --j.
>
> I still am trying to have 3.2.0 running on Win32 do: perl Makefile.PL &&
> make && make test for buildbot so far getting tons of errors. Cygwin
> builds fine. It has to include a whole bunch of different enviroment
> variables--thinking I could add them into my system settings? Anyway...
>
> Not sure if this is a show stopper for releasing 3.2.0 but thought I'd
> mention it. I'm in contact with someone who uses SA on a Win2003 server
> and I think he will be helpful in tracking down what is happening.
Well, we're kind of in a hurry at this stage ;)
I'd be inclined to release 3.2.0 without passing native-win32 tests, and
possibly with the win32 MISSING_HB_SEP bug -- purely because we can't
hold it up too much waiting for them to be investigated.
--j.
