Re: Upcoming back-branch releases

We're about due for a new set of back-branch update releases. After
some discussion among the packagers, it seems the best window for
getting this done before the holiday season sets in is next week.

Also, as previously mentioned, we're experimenting with weekday
rather than over-the-weekend release windows, since this is much
more convenient for Dave Page and his crew at EDB.

The current plan is to wrap tarballs on Monday Dec 3 for public
announcement Thursday Dec 6.

regards, tom lane

On Mon, Nov 26, 2012 at 9:17 PM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:

> We're about due for a new set of back-branch update releases. After
> some discussion among the packagers, it seems the best window for
> getting this done before the holiday season sets in is next week.
>
> Also, as previously mentioned, we're experimenting with weekday
> rather than over-the-weekend release windows, since this is much
> more convenient for Dave Page and his crew at EDB.
>
> The current plan is to wrap tarballs on Monday Dec 3 for public
> announcement Thursday Dec 6.
>
>
There is one open bug reported by Amit Kapila that I feel is important. I'd
investigated this and Simon had volunteered to fix the same.
http://archives.postgresql.org/pgsql-hackers/2012-10/msg01584.php

I'm not sure about the project policies about fixing known bugs before
minor releases, but AFAICS this one can lead to corrupt indexes and as a
consequence return wrong query results (as shown in the case by Amit). So
IMHO we should fix this.

Thanks,
Pavan

On 2012-11-26 21:35:05 +0530, Pavan Deolasee wrote:
> On Mon, Nov 26, 2012 at 9:17 PM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
>
> > We're about due for a new set of back-branch update releases. After
> > some discussion among the packagers, it seems the best window for
> > getting this done before the holiday season sets in is next week.
> >
> > Also, as previously mentioned, we're experimenting with weekday
> > rather than over-the-weekend release windows, since this is much
> > more convenient for Dave Page and his crew at EDB.
> >
> > The current plan is to wrap tarballs on Monday Dec 3 for public
> > announcement Thursday Dec 6.
> >
> >
> There is one open bug reported by Amit Kapila that I feel is important. I'd
> investigated this and Simon had volunteered to fix the same.
> http://archives.postgresql.org/pgsql-hackers/2012-10/msg01584.php

I have submitted a proposed fix for it friday:

http://archives.postgresql.org/message-id/20121124155005.GA10299%40awork2.anarazel.de

For some reason the web ui only shows one of the the attachements
though...

They also are in:
http://git.postgresql.org/gitweb/?p=users/andresfreund/postgres.git;a=shortlog;h=refs/heads/bugfixes
Greetings,

Andres Freund

--
Andres Freund http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services

On Mon, Nov 26, 2012 at 5:21 PM, Andres Freund <andres(at)2ndquadrant(dot)com> wrote:
> On 2012-11-26 21:35:05 +0530, Pavan Deolasee wrote:
>> On Mon, Nov 26, 2012 at 9:17 PM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
>>
>> > We're about due for a new set of back-branch update releases. After
>> > some discussion among the packagers, it seems the best window for
>> > getting this done before the holiday season sets in is next week.
>> >
>> > Also, as previously mentioned, we're experimenting with weekday
>> > rather than over-the-weekend release windows, since this is much
>> > more convenient for Dave Page and his crew at EDB.
>> >
>> > The current plan is to wrap tarballs on Monday Dec 3 for public
>> > announcement Thursday Dec 6.
>> >
>> >
>> There is one open bug reported by Amit Kapila that I feel is important. I'd
>> investigated this and Simon had volunteered to fix the same.
>> http://archives.postgresql.org/pgsql-hackers/2012-10/msg01584.php
>
> I have submitted a proposed fix for it friday:
>
> http://archives.postgresql.org/message-id/20121124155005.GA10299%40awork2.anarazel.de
>
> For some reason the web ui only shows one of the the attachements
> though...

Did you by any chance use git-send-email to send it?

That one is known to confuse how majordomo sticks it in the mbox files
that are then later used to generate the archives...

(Yes, a better way to deal with that is in the works. But it would
certainly help to know if that's how the email was created, so we have
more datapoints)

--
Magnus Hagander
Me: http://www.hagander.net/
Work: http://www.redpill-linpro.com/

On 2012-11-26 17:27:11 +0100, Magnus Hagander wrote:
> On Mon, Nov 26, 2012 at 5:21 PM, Andres Freund <andres(at)2ndquadrant(dot)com> wrote:
> > On 2012-11-26 21:35:05 +0530, Pavan Deolasee wrote:
> >> On Mon, Nov 26, 2012 at 9:17 PM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> >>
> >> > We're about due for a new set of back-branch update releases. After
> >> > some discussion among the packagers, it seems the best window for
> >> > getting this done before the holiday season sets in is next week.
> >> >
> >> > Also, as previously mentioned, we're experimenting with weekday
> >> > rather than over-the-weekend release windows, since this is much
> >> > more convenient for Dave Page and his crew at EDB.
> >> >
> >> > The current plan is to wrap tarballs on Monday Dec 3 for public
> >> > announcement Thursday Dec 6.
> >> >
> >> >
> >> There is one open bug reported by Amit Kapila that I feel is important. I'd
> >> investigated this and Simon had volunteered to fix the same.
> >> http://archives.postgresql.org/pgsql-hackers/2012-10/msg01584.php
> >
> > I have submitted a proposed fix for it friday:
> >
> > http://archives.postgresql.org/message-id/20121124155005.GA10299%40awork2.anarazel.de
> >
> > For some reason the web ui only shows one of the the attachements
> > though...
>
> Did you by any chance use git-send-email to send it?
>
> That one is known to confuse how majordomo sticks it in the mbox files
> that are then later used to generate the archives...
>
> (Yes, a better way to deal with that is in the works. But it would
> certainly help to know if that's how the email was created, so we have
> more datapoints)

No git-send-email in this case (although I used it in the past and
likely will use it again). Mutt instead. The attachements were generated
by "git format-patch HEAD~2" and then normally attached to the
email. format-patch is internally used by send-email though, so that
might be a hint...

Greetings,

Andres Freund

--
Andres Freund http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services

Andres Freund escribió:
> On 2012-11-26 17:27:11 +0100, Magnus Hagander wrote:
> > On Mon, Nov 26, 2012 at 5:21 PM, Andres Freund <andres(at)2ndquadrant(dot)com> wrote:

> > > I have submitted a proposed fix for it friday:
> > >
> > > http://archives.postgresql.org/message-id/20121124155005.GA10299%40awork2.anarazel.de
> > >
> > > For some reason the web ui only shows one of the the attachements
> > > though...
> >
> > Did you by any chance use git-send-email to send it?
> >
> > That one is known to confuse how majordomo sticks it in the mbox files
> > that are then later used to generate the archives...
> >
> > (Yes, a better way to deal with that is in the works. But it would
> > certainly help to know if that's how the email was created, so we have
> > more datapoints)
>
> No git-send-email in this case (although I used it in the past and
> likely will use it again). Mutt instead. The attachements were generated
> by "git format-patch HEAD~2" and then normally attached to the
> email. format-patch is internally used by send-email though, so that
> might be a hint...

The problem is that those files start with the infamous "^From " header
line which Mhonarc is so old-fashioned about, so the patches ended up as
separate emails. Note the attachment in the email linked above is
empty.

You can see the patches here:
http://archives.postgresql.org/pgsql-hackers/2012-11/msg01278.php
http://archives.postgresql.org/pgsql-hackers/2012-11/msg01279.php

--
Álvaro Herrera http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services