Re: ERROR: Failed to build any 5-way joins

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: "Magnus Hagander" <magnus(at)hagander(dot)net>
Cc: smarlowe(at)g2switchworks(dot)com, rod(at)iol(dot)ie, pgsql-general(at)postgresql(dot)org
Subject: Re: ERROR: Failed to build any 5-way joins
Date: 2007-04-24 05:59:43
Message-ID: 24225.1177394383@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

"Magnus Hagander" <magnus(at)hagander(dot)net> writes:
>> If I remember correctly, after a few "where's this new release" snafus,
>> the order of the day now is to silently release the new version,
>> announcing it on -hackers only, until all the ftp mirrors have updated,
>> then to announce it publicly when all the copies have been made.

> correct. the wait is also for the official binary distributions to be packaged.

Yah. The actual process for the last few updates has involved wrapping
the "master source" tarballs on a Thursday evening with public
announcement the next Monday. This gives a couple of days for the RPM
and Windows packagers to do their thing, then another 24 hours or so
for the various mirrors to pick up the files before we announce. This
over-the-weekend schedule isn't set in concrete, but it seems to be
convenient for most of the packagers at the moment.

While there's no need to hide a simple bug-fix update, it's widely
considered important that security issues not become public before a fix
is available. This just-completed cycle was embarrassingly leaky, in
that there was a whole lot of unintended public evidence that a security
release was about to happen. We have all the policies and procedures in
place, but we seem to need a bit more practice at executing them...

regards, tom lane

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Richard Huxton 2007-04-24 07:45:21 Re: Setting table ids in slony
Previous Message Tino Wildenhain 2007-04-24 05:54:04 Re: Major differences between 7.1.1 and 7.4.x