On Tue, Jan 7, 2014 at 1:30 PM, Dan Robinson <[email protected]> wrote:
> This is looking like a nice upgrade.
>
> What does the release pipeline/process usually look like? Something to the
> effect of "keep releasing alphas until they're stable, then release a stable
> version"?

Yeah. Major new features will also go in a parallel branch at this
point, should one materialize, that will constitute v0.8.

I'm not sure what kinds of people are running the
Alpha/Beta/RC/Release.[0-n] so I'm guessing on how fast to proceed.  I
probably err towards the "too slow" side just because of the holidays
and some other uses of my time at the moment.

What would get me to accelerate this process would be positive reports
of people using the alpha and pretty good confidence that crucial
backup and restore mechanisms are working as expected, and perhaps
also that the 'delete' operators haven't developed a tendency to be
overzealous.  These are not regressions that have proved particularly
common, but as a disaster recovery program it would seem best to ward
off users that may have the temptation to install this new code
without testing the backup *and* restore: that's what I'm trying to
get across with the "Alpha" designation.

There are a few things I'd like to see before Beta, but I'm not
dead-set on these if it looks like it's not going to happen in a
timely manner:

* Documenting how to use Swift (but the number of backends is making
  the existing documentation format harder to deal with, so this is
  harder than it sounds to do well).

* I have been working on and off on some integration test stuff and
  I'd like to get it in a state to be committed.  It won't poke any of
  the WAL-E mechanism as seen in production in any significant way.

-- 
You received this message because you are subscribed to the Google Groups 
"wal-e" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/groups/opt_out.

Reply via email to