[bind10-dev] release not branched yet -- still in freeze

[Michael Graff]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 2010-06-02 3:56 PM, Jeremy C. Reed wrote:
> Note that the freeze is over. It was branched. The release has been 
> shipped.
> 
> As for the build problems:
> 
> - I may need to have a monitor that reports to me when an auto build 
> server disappears. (I still haven't heard from OPS.)
> 
> - Do I need more build systems doing "identical" autobuilds? I don't 
> think so since I already have a few different systems and not sure if 
> more reports on same problems helps or not ... but that goes with next 
> question...

I don't know what you mean by "identical"

It seems to me we should always have common branches compiling in more
than one spot.  I'm not certain we need 15 error reports though.

> - Do we need more frequent or louder reminders when builds/checks/ 
> installs fail?

No, please, not more.  Better.  When I see "build failed" I generally
don't spend time to run off and find out why; this is a problem I have
with BIND 9 builds as well.  So, AIX failed.  Oh well.

If it gave me a snippit of failed build, I'd care and likely fix it if
it was mine.  No more time wasted running off only to find out it was
code someone else committed.

> - Do we need some policies for how to handle these failures?

Beyond "fix them"?  It would be really nice if we had a way to ensure
people don't commit code that won't compile.  Checks are there for a
reason, and should not just be ignored.  But, do we need a policy beyond
"do a full trial compile" and "Fix check-time bugs you introduce
quickly"?  No, I don't think so.  We do reserve the right to make a mark
next to your name for "breaking the build" though.  :)

- --Michael
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.8 (Darwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAkwGyCYACgkQ+NNi0s9NRJ1ACwCgljvwMkVWroH1NQxeZsxKrXe/
/a0AoJe4bBACe2GxqeHLGVId3SZZiez4
=UcJB
-----END PGP SIGNATURE-----