pgsql: docs: update release notes for 8.4.18, 9.0.14, 9.1.10, 9.2.5, 9

docs: update release notes for 8.4.18, 9.0.14, 9.1.10, 9.2.5, 9.3.1

Branch
------
REL9_2_STABLE

Details
-------
http://git.postgresql.org/pg/commitdiff/df9ede9063915bca311a14a482d97ef51d2082d3

Modified Files
--------------
doc/src/sgml/release-8.4.sgml | 193 ++++++++++++++++++++
doc/src/sgml/release-9.0.sgml | 257 ++++++++++++++++++++++++++
doc/src/sgml/release-9.1.sgml | 321 +++++++++++++++++++++++++++++++++
doc/src/sgml/release-9.2.sgml | 397 +++++++++++++++++++++++++++++++++++++++++
4 files changed, 1168 insertions(+)

(2013/10/08 10:35), Bruce Momjian wrote:
> docs: update release notes for 8.4.18, 9.0.14, 9.1.10, 9.2.5, 9.3.1
Thank you for createing good release note. I have one comment.

In 9.1 and 9.2 release note, Is "Improve WAL segment timeline handling during
recovery" means commit which is "Install recycled WAL segments with current
timeline ID during recovery"? This is not so simple problem fix.
This bug caused failing PITR which is finished archive recovery on the way. When
it occered, it seemed to finish archive recovery without problem.

I think it is comparatively big problem, so we should write it in release note.
Please fix it under following.

+ <listitem>
+ <para>
+ Fix WAL segment timeline handling during recovery (Mitsumasa
+ KONDO, Heikki Linnakangas)
+ </para>
+
+ <para>
+ When target timeline is up and executing restart point in archive recovery
+ mode, archive recovery is failed on the way, because failing recycle of +
WAL. When this problem occurred, it seemed to finish success of archive +
recovery without problem.
+ </para>
+ </listitem>

Regards,
--
Mitsumasa KONDO
NTT Open Source Software Center

On Tue, Oct 8, 2013 at 01:25:30PM +0900, KONDO Mitsumasa wrote:
> (2013/10/08 10:35), Bruce Momjian wrote:
> >docs: update release notes for 8.4.18, 9.0.14, 9.1.10, 9.2.5, 9.3.1
> Thank you for creating good release note. I have one comment.
>
> In 9.1 and 9.2 release note, Is "Improve WAL segment timeline
> handling during recovery" means commit which is "Install recycled
> WAL segments with current timeline ID during recovery"? This is not
> so simple problem fix.
> This bug caused failing PITR which is finished archive recovery on
> the way. When it occered, it seemed to finish archive recovery
> without problem.
>
> I think it is comparatively big problem, so we should write it in release note.
> Please fix it under following.
>
> + <listitem>
> + <para>
> + Fix WAL segment timeline handling during recovery (Mitsumasa
> + KONDO, Heikki Linnakangas)
> + </para>
> +
> + <para>
> + When target timeline is up and executing restart point in archive recovery
> + mode, archive recovery is failed on the way, because failing
> recycle of + WAL. When this problem occurred, it seemed to finish
> success of archive + recovery without problem.
> + </para>
> + </listitem>

First, I want to apologize for not completing the release notes earlier
so that others could review them. I started working on the release
notes on Friday, but my unfamiliarity with the process and fear of
making a mistake caused many delays. I have improved the documentation
on the process which will hopefully help next time.

Second, I have read the thread beind this patch:

http://www.postgresql.org/message-id/flat/51798552(dot)2010102(at)vmware(dot)com#51798552(dot)2010102@vmware.com

You are right that there is alot of details skipped in the release note
text. I have developed the attached patch which I think does a better
job. Is it OK?

--
Bruce Momjian <bruce(at)momjian(dot)us> http://momjian.us
EnterpriseDB http://enterprisedb.com

+ It's impossible for everything to be true. +

Sorry for my reply late...

(2013/10/08 23:26), Bruce Momjian wrote:
> First, I want to apologize for not completing the release notes earlier
> so that others could review them. I started working on the release
> notes on Friday, but my unfamiliarity with the process and fear of
> making a mistake caused many delays. I have improved the documentation
> on the process which will hopefully help next time.
There isn't anything in particular that I was dissatisfied about it.

> You are right that there is alot of details skipped in the release note
> text. I have developed the attached patch which I think does a better
> job. Is it OK?
Yes, off course!
Thanks for your sincere action!

Regards,
--
Mitsumasa KONDO
NTT Open Source Software Center

On Tue, Oct 15, 2013 at 02:32:47PM +0900, KONDO Mitsumasa wrote:
> Sorry for my reply late...
>
> (2013/10/08 23:26), Bruce Momjian wrote:
> > First, I want to apologize for not completing the release notes earlier
> > so that others could review them. I started working on the release
> > notes on Friday, but my unfamiliarity with the process and fear of
> > making a mistake caused many delays. I have improved the documentation
> > on the process which will hopefully help next time.
> There isn't anything in particular that I was dissatisfied about it.
>
> >You are right that there is alot of details skipped in the release note
> >text. I have developed the attached patch which I think does a better
> >job. Is it OK?
> Yes, off course!
> Thanks for your sincere action!

Thanks, patch applied back through 9.1.

--
Bruce Momjian <bruce(at)momjian(dot)us> http://momjian.us
EnterpriseDB http://enterprisedb.com

+ Everyone has their own god. +