database/postgres
1
0
Fork 0
mirror of https://github.com/postgres/postgres.git synced 2025-07-28 23:42:10 +03:00
Code Activity

Release notes for 9.5.3, 9.4.8, 9.3.13, 9.2.17, 9.1.22.

Browse Source
This commit is contained in:
Tom Lane
2016-05-07 17:26:24 -04:00
parent 673bceaf3f
commit 828fe49148
5 changed files with 1357 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

165
doc/src/sgml/release-9.1.sgml
View File

@ -1,6 +1,171 @@
<!-- doc/src/sgml/release-9.1.sgml -->
<!-- See header comment in release.sgml about typical markup -->
<sect1 id="release-9-1-22">
<title>Release 9.1.22</title>
<note>
<title>Release Date</title>
<simpara>2016-05-12</simpara>
</note>
<para>
This release contains a variety of fixes from 9.1.21.
For information about new features in the 9.1 major release, see
<xref linkend="release-9-1">.
</para>
<para>
The <productname>PostgreSQL</> community will stop releasing updates
for the 9.1.X release series in September 2016.
Users are encouraged to update to a newer release branch soon.
</para>
<sect2>
<title>Migration to Version 9.1.22</title>
<para>
A dump/restore is not required for those running 9.1.X.
</para>
<para>
However, if you are upgrading from a version earlier than 9.1.16,
see <xref linkend="release-9-1-16">.
</para>
</sect2>
<sect2>
<title>Changes</title>
<itemizedlist>
<listitem>
<para>
Clear the OpenSSL error queue before OpenSSL calls, rather than
assuming it's clear already; and make sure we leave it clear
afterwards (Peter Geoghegan, Dave Vitek, Peter Eisentraut)
</para>
<para>
This change prevents problems when there are multiple connections
using OpenSSL within a single process and not all the code involved
follows the same rules for when to clear the error queue.
Failures have been reported specifically when a client application
uses SSL connections in <application>libpq</> concurrently with
SSL connections using the PHP, Python, or Ruby wrappers for OpenSSL.
It's possible for similar problems to arise within the server as well,
if an extension module establishes an outgoing SSL connection.
</para>
</listitem>
<listitem>
<para>
Fix <quote>failed to build any <replaceable>N</>-way joins</quote>
planner error with a full join enclosed in the right-hand side of a
left join (Tom Lane)
</para>
</listitem>
<listitem>
<para>
Fix possible misbehavior of <literal>TH</>, <literal>th</>,
and <literal>Y,YYY</> format codes in <function>to_timestamp()</>
(Tom Lane)
</para>
<para>
These could advance off the end of the input string, causing subsequent
format codes to read garbage.
</para>
</listitem>
<listitem>
<para>
Fix dumping of rules and views in which the <replaceable>array</>
argument of a <literal><replaceable>value</> <replaceable>operator</>
ANY (<replaceable>array</>)</literal> construct is a sub-SELECT
(Tom Lane)
</para>
</listitem>
<listitem>
<para>
Make <application>pg_regress</> use a startup timeout from the
<envar>PGCTLTIMEOUT</> environment variable, if that's set (Tom Lane)
</para>
<para>
This is for consistency with a behavior recently added
to <application>pg_ctl</>; it eases automated testing on slow machines.
</para>
</listitem>
<listitem>
<para>
Fix <application>pg_upgrade</> to correctly restore extension
membership for operator families containing only one operator class
(Tom Lane)
</para>
<para>
In such a case, the operator family was restored into the new database,
but it was no longer marked as part of the extension. This had no
immediate ill effects, but would cause later <application>pg_dump</>
runs to emit output that would cause (harmless) errors on restore.
</para>
</listitem>
<listitem>
<para>
Rename internal function <function>strtoi()</>
to <function>strtoint()</> to avoid conflict with a NetBSD library
function (Thomas Munro)
</para>
</listitem>
<listitem>
<para>
Fix reporting of errors from <function>bind()</>
and <function>listen()</> system calls on Windows (Tom Lane)
</para>
</listitem>
<listitem>
<para>
Reduce verbosity of compiler output when building with Microsoft Visual
Studio (Christian Ullrich)
</para>
</listitem>
<listitem>
<para>
Avoid possibly-unsafe use of Windows' <function>FormatMessage()</>
function (Christian Ullrich)
</para>
<para>
Use the <literal>FORMAT_MESSAGE_IGNORE_INSERTS</> flag where
appropriate. No live bug is known to exist here, but it seems like a
good idea to be careful.
</para>
</listitem>
<listitem>
<para>
Update time zone data files to <application>tzdata</> release 2016d
for DST law changes in Russia and Venezuela. There are new zone
names <literal>Europe/Kirov</> and <literal>Asia/Tomsk</> to reflect
the fact that these regions now have different time zone histories from
adjacent regions.
</para>
</listitem>
</itemizedlist>
</sect2>
</sect1>
<sect1 id="release-9-1-21">
<title>Release 9.1.21</title>

196
doc/src/sgml/release-9.2.sgml
View File

@ -1,6 +1,202 @@
<!-- doc/src/sgml/release-9.2.sgml -->
<!-- See header comment in release.sgml about typical markup -->
<sect1 id="release-9-2-17">
<title>Release 9.2.17</title>
<note>
<title>Release Date</title>
<simpara>2016-05-12</simpara>
</note>
<para>
This release contains a variety of fixes from 9.2.16.
For information about new features in the 9.2 major release, see
<xref linkend="release-9-2">.
</para>
<sect2>
<title>Migration to Version 9.2.17</title>
<para>
A dump/restore is not required for those running 9.2.X.
</para>
<para>
However, if you are upgrading from a version earlier than 9.2.11,
see <xref linkend="release-9-2-11">.
</para>
</sect2>
<sect2>
<title>Changes</title>
<itemizedlist>
<listitem>
<para>
Clear the OpenSSL error queue before OpenSSL calls, rather than
assuming it's clear already; and make sure we leave it clear
afterwards (Peter Geoghegan, Dave Vitek, Peter Eisentraut)
</para>
<para>
This change prevents problems when there are multiple connections
using OpenSSL within a single process and not all the code involved
follows the same rules for when to clear the error queue.
Failures have been reported specifically when a client application
uses SSL connections in <application>libpq</> concurrently with
SSL connections using the PHP, Python, or Ruby wrappers for OpenSSL.
It's possible for similar problems to arise within the server as well,
if an extension module establishes an outgoing SSL connection.
</para>
</listitem>
<listitem>
<para>
Fix <quote>failed to build any <replaceable>N</>-way joins</quote>
planner error with a full join enclosed in the right-hand side of a
left join (Tom Lane)
</para>
</listitem>
<listitem>
<para>
Fix incorrect handling of equivalence-class tests in multilevel
nestloop plans (Tom Lane)
</para>
<para>
Given a three-or-more-way equivalence class of variables, such
as <literal>X.X = Y.Y = Z.Z</>, it was possible for the planner to omit
some of the tests needed to enforce that all the variables are actually
equal, leading to join rows being output that didn't satisfy
the <literal>WHERE</> clauses. For various reasons, erroneous plans
were seldom selected in practice, so that this bug has gone undetected
for a long time.
</para>
</listitem>
<listitem>
<para>
Fix possible misbehavior of <literal>TH</>, <literal>th</>,
and <literal>Y,YYY</> format codes in <function>to_timestamp()</>
(Tom Lane)
</para>
<para>
These could advance off the end of the input string, causing subsequent
format codes to read garbage.
</para>
</listitem>
<listitem>
<para>
Fix dumping of rules and views in which the <replaceable>array</>
argument of a <literal><replaceable>value</> <replaceable>operator</>
ANY (<replaceable>array</>)</literal> construct is a sub-SELECT
(Tom Lane)
</para>
</listitem>
<listitem>
<para>
Make <application>pg_regress</> use a startup timeout from the
<envar>PGCTLTIMEOUT</> environment variable, if that's set (Tom Lane)
</para>
<para>
This is for consistency with a behavior recently added
to <application>pg_ctl</>; it eases automated testing on slow machines.
</para>
</listitem>
<listitem>
<para>
Fix <application>pg_upgrade</> to correctly restore extension
membership for operator families containing only one operator class
(Tom Lane)
</para>
<para>
In such a case, the operator family was restored into the new database,
but it was no longer marked as part of the extension. This had no
immediate ill effects, but would cause later <application>pg_dump</>
runs to emit output that would cause (harmless) errors on restore.
</para>
</listitem>
<listitem>
<para>
Back-port 9.4-era memory-barrier code changes into 9.2 and 9.3 (Tom Lane)
</para>
<para>
These changes were not originally needed in pre-9.4 branches, but we
recently back-patched a fix that expected the barrier code to work
properly. Only IA64 (when using icc), HPPA, and Alpha platforms are
affected.
</para>
</listitem>
<listitem>
<para>
Reduce the number of SysV semaphores used by a build configured with
<option>--disable-spinlocks</> (Tom Lane)
</para>
</listitem>
<listitem>
<para>
Rename internal function <function>strtoi()</>
to <function>strtoint()</> to avoid conflict with a NetBSD library
function (Thomas Munro)
</para>
</listitem>
<listitem>
<para>
Fix reporting of errors from <function>bind()</>
and <function>listen()</> system calls on Windows (Tom Lane)
</para>
</listitem>
<listitem>
<para>
Reduce verbosity of compiler output when building with Microsoft Visual
Studio (Christian Ullrich)
</para>
</listitem>
<listitem>
<para>
Avoid possibly-unsafe use of Windows' <function>FormatMessage()</>
function (Christian Ullrich)
</para>
<para>
Use the <literal>FORMAT_MESSAGE_IGNORE_INSERTS</> flag where
appropriate. No live bug is known to exist here, but it seems like a
good idea to be careful.
</para>
</listitem>
<listitem>
<para>
Update time zone data files to <application>tzdata</> release 2016d
for DST law changes in Russia and Venezuela. There are new zone
names <literal>Europe/Kirov</> and <literal>Asia/Tomsk</> to reflect
the fact that these regions now have different time zone histories from
adjacent regions.
</para>
</listitem>
</itemizedlist>
</sect2>
</sect1>
<sect1 id="release-9-2-16">
<title>Release 9.2.16</title>

232
doc/src/sgml/release-9.3.sgml
View File

@ -1,6 +1,236 @@
<!-- doc/src/sgml/release-9.3.sgml -->
<!-- See header comment in release.sgml about typical markup -->
<sect1 id="release-9-3-13">
<title>Release 9.3.13</title>
<note>
<title>Release Date</title>
<simpara>2016-05-12</simpara>
</note>
<para>
This release contains a variety of fixes from 9.3.12.
For information about new features in the 9.3 major release, see
<xref linkend="release-9-3">.
</para>
<sect2>
<title>Migration to Version 9.3.13</title>
<para>
A dump/restore is not required for those running 9.3.X.
</para>
<para>
However, if you are upgrading from a version earlier than 9.3.9,
see <xref linkend="release-9-3-9">.
</para>
</sect2>
<sect2>
<title>Changes</title>
<itemizedlist>
<listitem>
<para>
Clear the OpenSSL error queue before OpenSSL calls, rather than
assuming it's clear already; and make sure we leave it clear
afterwards (Peter Geoghegan, Dave Vitek, Peter Eisentraut)
</para>
<para>
This change prevents problems when there are multiple connections
using OpenSSL within a single process and not all the code involved
follows the same rules for when to clear the error queue.
Failures have been reported specifically when a client application
uses SSL connections in <application>libpq</> concurrently with
SSL connections using the PHP, Python, or Ruby wrappers for OpenSSL.
It's possible for similar problems to arise within the server as well,
if an extension module establishes an outgoing SSL connection.
</para>
</listitem>
<listitem>
<para>
Fix <quote>failed to build any <replaceable>N</>-way joins</quote>
planner error with a full join enclosed in the right-hand side of a
left join (Tom Lane)
</para>
</listitem>
<listitem>
<para>
Fix incorrect handling of equivalence-class tests in multilevel
nestloop plans (Tom Lane)
</para>
<para>
Given a three-or-more-way equivalence class of variables, such
as <literal>X.X = Y.Y = Z.Z</>, it was possible for the planner to omit
some of the tests needed to enforce that all the variables are actually
equal, leading to join rows being output that didn't satisfy
the <literal>WHERE</> clauses. For various reasons, erroneous plans
were seldom selected in practice, so that this bug has gone undetected
for a long time.
</para>
</listitem>
<listitem>
<para>
Fix possible misbehavior of <literal>TH</>, <literal>th</>,
and <literal>Y,YYY</> format codes in <function>to_timestamp()</>
(Tom Lane)
</para>
<para>
These could advance off the end of the input string, causing subsequent
format codes to read garbage.
</para>
</listitem>
<listitem>
<para>
Fix dumping of rules and views in which the <replaceable>array</>
argument of a <literal><replaceable>value</> <replaceable>operator</>
ANY (<replaceable>array</>)</literal> construct is a sub-SELECT
(Tom Lane)
</para>
</listitem>
<listitem>
<para>
Make <application>pg_regress</> use a startup timeout from the
<envar>PGCTLTIMEOUT</> environment variable, if that's set (Tom Lane)
</para>
<para>
This is for consistency with a behavior recently added
to <application>pg_ctl</>; it eases automated testing on slow machines.
</para>
</listitem>
<listitem>
<para>
Fix <application>pg_upgrade</> to correctly restore extension
membership for operator families containing only one operator class
(Tom Lane)
</para>
<para>
In such a case, the operator family was restored into the new database,
but it was no longer marked as part of the extension. This had no
immediate ill effects, but would cause later <application>pg_dump</>
runs to emit output that would cause (harmless) errors on restore.
</para>
</listitem>
<listitem>
<para>
Fix <application>pg_upgrade</> to not fail when new-cluster TOAST rules
differ from old (Tom Lane)
</para>
<para>
<application>pg_upgrade</> had special-case code to handle the
situation where the new <productname>PostgreSQL</> version thinks that
a table should have a TOAST table while the old version did not. That
code was broken, so remove it, and instead do nothing in such cases;
there seems no reason to believe that we can't get along fine without
a TOAST table if that was okay according to the old version's rules.
</para>
</listitem>
<listitem>
<!--
Author: Tom Lane <tgl@sss.pgh.pa.us>
Branch: REL9_3_STABLE [f4f4f6990] 2016-04-15 16:49:48 -0400
Branch: REL9_2_STABLE [d7dbc882d] 2016-04-15 16:49:48 -0400
Author: Tom Lane <tgl@sss.pgh.pa.us>
Branch: REL9_3_STABLE [992df9658] 2016-04-16 10:42:07 -0400
Branch: REL9_2_STABLE [9008922bf] 2016-04-16 10:41:57 -0400
Author: Tom Lane <tgl@sss.pgh.pa.us>
Branch: REL9_3_STABLE [35166fd76] 2016-04-18 13:19:52 -0400
Branch: REL9_2_STABLE [37f30b251] 2016-04-18 13:19:52 -0400
-->
<para>
Back-port 9.4-era memory-barrier code changes into 9.2 and 9.3 (Tom Lane)
</para>
<para>
These changes were not originally needed in pre-9.4 branches, but we
recently back-patched a fix that expected the barrier code to work
properly. Only IA64 (when using icc), HPPA, and Alpha platforms are
affected.
</para>
</listitem>
<listitem>
<para>
Reduce the number of SysV semaphores used by a build configured with
<option>--disable-spinlocks</> (Tom Lane)
</para>
</listitem>
<listitem>
<para>
Rename internal function <function>strtoi()</>
to <function>strtoint()</> to avoid conflict with a NetBSD library
function (Thomas Munro)
</para>
</listitem>
<listitem>
<para>
Fix reporting of errors from <function>bind()</>
and <function>listen()</> system calls on Windows (Tom Lane)
</para>
</listitem>
<listitem>
<para>
Reduce verbosity of compiler output when building with Microsoft Visual
Studio (Christian Ullrich)
</para>
</listitem>
<listitem>
<para>
Fix <function>putenv()</> to work properly with Visual Studio 2013
(Michael Paquier)
</para>
</listitem>
<listitem>
<para>
Avoid possibly-unsafe use of Windows' <function>FormatMessage()</>
function (Christian Ullrich)
</para>
<para>
Use the <literal>FORMAT_MESSAGE_IGNORE_INSERTS</> flag where
appropriate. No live bug is known to exist here, but it seems like a
good idea to be careful.
</para>
</listitem>
<listitem>
<para>
Update time zone data files to <application>tzdata</> release 2016d
for DST law changes in Russia and Venezuela. There are new zone
names <literal>Europe/Kirov</> and <literal>Asia/Tomsk</> to reflect
the fact that these regions now have different time zone histories from
adjacent regions.
</para>
</listitem>
</itemizedlist>
</sect2>
</sect1>
<sect1 id="release-9-3-12">
<title>Release 9.3.12</title>
@ -9689,7 +9919,7 @@ ALTER EXTENSION hstore UPDATE;
<listitem>
<para>
Remove <filename>src/tool/backend</>, now that the content is on
Remove <filename>src/tools/backend</>, now that the content is on
the <productname>PostgreSQL</> wiki (Bruce Momjian)
</para>
</listitem>

263
doc/src/sgml/release-9.4.sgml
View File

@ -1,6 +1,269 @@
<!-- doc/src/sgml/release-9.4.sgml -->
<!-- See header comment in release.sgml about typical markup -->
<sect1 id="release-9-4-8">
<title>Release 9.4.8</title>
<note>
<title>Release Date</title>
<simpara>2016-05-12</simpara>
</note>
<para>
This release contains a variety of fixes from 9.4.7.
For information about new features in the 9.4 major release, see
<xref linkend="release-9-4">.
</para>
<sect2>
<title>Migration to Version 9.4.8</title>
<para>
A dump/restore is not required for those running 9.4.X.
</para>
<para>
However, if you are upgrading from a version earlier than 9.4.6,
see <xref linkend="release-9-4-6">.
</para>
</sect2>
<sect2>
<title>Changes</title>
<itemizedlist>
<listitem>
<para>
Clear the OpenSSL error queue before OpenSSL calls, rather than
assuming it's clear already; and make sure we leave it clear
afterwards (Peter Geoghegan, Dave Vitek, Peter Eisentraut)
</para>
<para>
This change prevents problems when there are multiple connections
using OpenSSL within a single process and not all the code involved
follows the same rules for when to clear the error queue.
Failures have been reported specifically when a client application
uses SSL connections in <application>libpq</> concurrently with
SSL connections using the PHP, Python, or Ruby wrappers for OpenSSL.
It's possible for similar problems to arise within the server as well,
if an extension module establishes an outgoing SSL connection.
</para>
</listitem>
<listitem>
<para>
Fix <quote>failed to build any <replaceable>N</>-way joins</quote>
planner error with a full join enclosed in the right-hand side of a
left join (Tom Lane)
</para>
</listitem>
<listitem>
<para>
Fix incorrect handling of equivalence-class tests in multilevel
nestloop plans (Tom Lane)
</para>
<para>
Given a three-or-more-way equivalence class of variables, such
as <literal>X.X = Y.Y = Z.Z</>, it was possible for the planner to omit
some of the tests needed to enforce that all the variables are actually
equal, leading to join rows being output that didn't satisfy
the <literal>WHERE</> clauses. For various reasons, erroneous plans
were seldom selected in practice, so that this bug has gone undetected
for a long time.
</para>
</listitem>
<listitem>
<para>
Fix query-lifespan memory leak in GIN index scans (Julien Rouhaud)
</para>
</listitem>
<listitem>
<para>
Fix query-lifespan memory leak and potential index corruption hazard in
GIN index insertion (Tom Lane)
</para>
<para>
The memory leak would typically not amount to much in simple queries,
but it could be very substantial during a large GIN index build with
high <varname>maintenance_work_mem</>.
</para>
</listitem>
<listitem>
<para>
Fix possible misbehavior of <literal>TH</>, <literal>th</>,
and <literal>Y,YYY</> format codes in <function>to_timestamp()</>
(Tom Lane)
</para>
<para>
These could advance off the end of the input string, causing subsequent
format codes to read garbage.
</para>
</listitem>
<listitem>
<para>
Fix dumping of rules and views in which the <replaceable>array</>
argument of a <literal><replaceable>value</> <replaceable>operator</>
ANY (<replaceable>array</>)</literal> construct is a sub-SELECT
(Tom Lane)
</para>
</listitem>
<listitem>
<para>
Disallow newlines in <command>ALTER SYSTEM</> parameter values
(Tom Lane)
</para>
<para>
The configuration-file parser doesn't support embedded newlines in
string literals, so we mustn't allow them in values to be inserted
by <command>ALTER SYSTEM</>.
</para>
</listitem>
<listitem>
<para>
Fix <command>ALTER TABLE ... REPLICA IDENTITY USING INDEX</> to
work properly if an index on OID is selected (David Rowley)
</para>
</listitem>
<listitem>
<para>
Fix crash in logical decoding on alignment-picky platforms (Tom Lane,
Andres Freund)
</para>
<para>
The failure occurred only with a transaction large enough to spill to
disk and a primary-key change within that transaction.
</para>
</listitem>
<listitem>
<para>
Avoid repeated requests for feedback from receiver while shutting down
walsender (Nick Cleaton)
</para>
</listitem>
<listitem>
<para>
Make <application>pg_regress</> use a startup timeout from the
<envar>PGCTLTIMEOUT</> environment variable, if that's set (Tom Lane)
</para>
<para>
This is for consistency with a behavior recently added
to <application>pg_ctl</>; it eases automated testing on slow machines.
</para>
</listitem>
<listitem>
<para>
Fix <application>pg_upgrade</> to correctly restore extension
membership for operator families containing only one operator class
(Tom Lane)
</para>
<para>
In such a case, the operator family was restored into the new database,
but it was no longer marked as part of the extension. This had no
immediate ill effects, but would cause later <application>pg_dump</>
runs to emit output that would cause (harmless) errors on restore.
</para>
</listitem>
<listitem>
<para>
Fix <application>pg_upgrade</> to not fail when new-cluster TOAST rules
differ from old (Tom Lane)
</para>
<para>
<application>pg_upgrade</> had special-case code to handle the
situation where the new <productname>PostgreSQL</> version thinks that
a table should have a TOAST table while the old version did not. That
code was broken, so remove it, and instead do nothing in such cases;
there seems no reason to believe that we can't get along fine without
a TOAST table if that was okay according to the old version's rules.
</para>
</listitem>
<listitem>
<para>
Reduce the number of SysV semaphores used by a build configured with
<option>--disable-spinlocks</> (Tom Lane)
</para>
</listitem>
<listitem>
<para>
Rename internal function <function>strtoi()</>
to <function>strtoint()</> to avoid conflict with a NetBSD library
function (Thomas Munro)
</para>
</listitem>
<listitem>
<para>
Fix reporting of errors from <function>bind()</>
and <function>listen()</> system calls on Windows (Tom Lane)
</para>
</listitem>
<listitem>
<para>
Reduce verbosity of compiler output when building with Microsoft Visual
Studio (Christian Ullrich)
</para>
</listitem>
<listitem>
<para>
Fix <function>putenv()</> to work properly with Visual Studio 2013
(Michael Paquier)
</para>
</listitem>
<listitem>
<para>
Avoid possibly-unsafe use of Windows' <function>FormatMessage()</>
function (Christian Ullrich)
</para>
<para>
Use the <literal>FORMAT_MESSAGE_IGNORE_INSERTS</> flag where
appropriate. No live bug is known to exist here, but it seems like a
good idea to be careful.
</para>
</listitem>
<listitem>
<para>
Update time zone data files to <application>tzdata</> release 2016d
for DST law changes in Russia and Venezuela. There are new zone
names <literal>Europe/Kirov</> and <literal>Asia/Tomsk</> to reflect
the fact that these regions now have different time zone histories from
adjacent regions.
</para>
</listitem>
</itemizedlist>
</sect2>
</sect1>
<sect1 id="release-9-4-7">
<title>Release 9.4.7</title>

503
doc/src/sgml/release-9.5.sgml
View File

@ -1,6 +1,507 @@
<!-- doc/src/sgml/release-9.5.sgml -->
<!-- See header comment in release.sgml about typical markup -->
<sect1 id="release-9-5-3">
<title>Release 9.5.3</title>
<note>
<title>Release Date</title>
<simpara>2016-05-12</simpara>
</note>
<para>
This release contains a variety of fixes from 9.5.2.
For information about new features in the 9.5 major release, see
<xref linkend="release-9-5">.
</para>
<sect2>
<title>Migration to Version 9.5.3</title>
<para>
A dump/restore is not required for those running 9.5.X.
</para>
<para>
However, if you are upgrading from a version earlier than 9.5.2,
see <xref linkend="release-9-5-2">.
</para>
</sect2>
<sect2>
<title>Changes</title>
<itemizedlist>
<listitem>
<!--
Author: Peter Eisentraut <peter_e@gmx.net>
Branch: master [7c7d4fdda] 2016-04-08 14:11:56 -0400
Branch: REL9_5_STABLE [a3c17b2af] 2016-05-06 23:47:42 -0400
Branch: REL9_4_STABLE [e3b14de9f] 2016-05-07 00:10:31 -0400
Branch: REL9_3_STABLE [a9d8644de] 2016-05-07 00:10:17 -0400
Branch: REL9_2_STABLE [e3a493ac0] 2016-05-07 00:05:39 -0400
Branch: REL9_1_STABLE [9b676fd49] 2016-05-07 00:09:37 -0400
-->
<para>
Clear the OpenSSL error queue before OpenSSL calls, rather than
assuming it's clear already; and make sure we leave it clear
afterwards (Peter Geoghegan, Dave Vitek, Peter Eisentraut)
</para>
<para>
This change prevents problems when there are multiple connections
using OpenSSL within a single process and not all the code involved
follows the same rules for when to clear the error queue.
Failures have been reported specifically when a client application
uses SSL connections in <application>libpq</> concurrently with
SSL connections using the PHP, Python, or Ruby wrappers for OpenSSL.
It's possible for similar problems to arise within the server as well,
if an extension module establishes an outgoing SSL connection.
</para>
</listitem>
<listitem>
<!--
Author: Tom Lane <tgl@sss.pgh.pa.us>
Branch: master [80f66a9ad] 2016-04-21 20:05:58 -0400
Branch: REL9_5_STABLE [94c685a7c] 2016-04-21 20:05:58 -0400
Branch: REL9_4_STABLE [3232c2427] 2016-04-21 20:05:58 -0400
Branch: REL9_3_STABLE [d9742ac46] 2016-04-21 20:05:58 -0400
Branch: REL9_2_STABLE [ad2d32b57] 2016-04-21 20:05:58 -0400
Branch: REL9_1_STABLE [6882dbd34] 2016-04-21 20:05:58 -0400
-->
<para>
Fix <quote>failed to build any <replaceable>N</>-way joins</quote>
planner error with a full join enclosed in the right-hand side of a
left join (Tom Lane)
</para>
</listitem>
<listitem>
<!--
Author: Tom Lane <tgl@sss.pgh.pa.us>
Branch: master [207d5a656] 2016-04-29 20:19:38 -0400
Branch: REL9_5_STABLE [f25d0619c] 2016-04-29 20:19:38 -0400
Branch: REL9_4_STABLE [72edc8ffe] 2016-04-29 20:19:38 -0400
Branch: REL9_3_STABLE [67349e5a8] 2016-04-29 20:19:38 -0400
Branch: REL9_2_STABLE [f02cb8c9a] 2016-04-29 20:19:38 -0400
-->
<para>
Fix incorrect handling of equivalence-class tests in multilevel
nestloop plans (Tom Lane)
</para>
<para>
Given a three-or-more-way equivalence class of variables, such
as <literal>X.X = Y.Y = Z.Z</>, it was possible for the planner to omit
some of the tests needed to enforce that all the variables are actually
equal, leading to join rows being output that didn't satisfy
the <literal>WHERE</> clauses. For various reasons, erroneous plans
were seldom selected in practice, so that this bug has gone undetected
for a long time.
</para>
</listitem>
<listitem>
<!--
Author: Tom Lane <tgl@sss.pgh.pa.us>
Branch: master [abb164655] 2016-04-21 23:17:36 -0400
Branch: REL9_5_STABLE [81deadd31] 2016-04-21 23:17:36 -0400
-->
<para>
Fix corner-case parser failures occurring
when <xref linkend="guc-operator-precedence-warning"> is turned on
(Tom Lane)
</para>
<para>
An example is that <literal>SELECT (ARRAY[])::text[]</> gave an error,
though it worked without the parentheses.
</para>
</listitem>
<listitem>
<!--
Author: Tom Lane <tgl@sss.pgh.pa.us>
Branch: master [f0e766bd7] 2016-04-15 00:02:26 -0400
Branch: REL9_5_STABLE [edd57377f] 2016-04-15 00:02:26 -0400
Branch: REL9_4_STABLE [0479eccdc] 2016-04-15 00:02:26 -0400
-->
<para>
Fix query-lifespan memory leak in GIN index scans (Julien Rouhaud)
</para>
</listitem>
<listitem>
<!--
Author: Tom Lane <tgl@sss.pgh.pa.us>
Branch: master [bde361fef] 2016-04-20 14:25:15 -0400
Branch: REL9_5_STABLE [be2754456] 2016-04-20 14:25:15 -0400
Branch: REL9_4_STABLE [ef35afa35] 2016-04-20 14:25:15 -0400
-->
<para>
Fix query-lifespan memory leak and potential index corruption hazard in
GIN index insertion (Tom Lane)
</para>
<para>
The memory leak would typically not amount to much in simple queries,
but it could be very substantial during a large GIN index build with
high <varname>maintenance_work_mem</>.
</para>
</listitem>
<listitem>
<!--
Author: Tom Lane <tgl@sss.pgh.pa.us>
Branch: master [d136d600f] 2016-05-06 12:09:20 -0400
Branch: REL9_5_STABLE [504af1fe0] 2016-05-06 12:09:20 -0400
Branch: REL9_4_STABLE [1180868d1] 2016-05-06 12:09:20 -0400
Branch: REL9_3_STABLE [462456d8d] 2016-05-06 12:09:20 -0400
Branch: REL9_2_STABLE [11247dd99] 2016-05-06 12:09:20 -0400
Branch: REL9_1_STABLE [7bad282c3] 2016-05-06 12:09:20 -0400
-->
<para>
Fix possible misbehavior of <literal>TH</>, <literal>th</>,
and <literal>Y,YYY</> format codes in <function>to_timestamp()</>
(Tom Lane)
</para>
<para>
These could advance off the end of the input string, causing subsequent
format codes to read garbage.
</para>
</listitem>
<listitem>
<!--
Author: Tom Lane <tgl@sss.pgh.pa.us>
Branch: master [1f7c85b82] 2016-04-21 14:20:30 -0400
Branch: REL9_5_STABLE [4b52cc289] 2016-04-21 14:20:18 -0400
Branch: REL9_4_STABLE [679c92238] 2016-04-21 14:20:18 -0400
Branch: REL9_3_STABLE [e5882f26b] 2016-04-21 14:20:18 -0400
Branch: REL9_2_STABLE [c7c145e4f] 2016-04-21 14:20:18 -0400
Branch: REL9_1_STABLE [663624e60] 2016-04-21 14:20:18 -0400
-->
<para>
Fix dumping of rules and views in which the <replaceable>array</>
argument of a <literal><replaceable>value</> <replaceable>operator</>
ANY (<replaceable>array</>)</literal> construct is a sub-SELECT
(Tom Lane)
</para>
</listitem>
<listitem>
<!--
Author: Tom Lane <tgl@sss.pgh.pa.us>
Branch: master [99f3b5613] 2016-04-04 18:05:23 -0400
Branch: REL9_5_STABLE [f3d17491c] 2016-04-04 18:05:23 -0400
Branch: REL9_4_STABLE [28148e258] 2016-04-04 18:05:24 -0400
-->
<para>
Disallow newlines in <command>ALTER SYSTEM</> parameter values
(Tom Lane)
</para>
<para>
The configuration-file parser doesn't support embedded newlines in
string literals, so we mustn't allow them in values to be inserted
by <command>ALTER SYSTEM</>.
</para>
</listitem>
<listitem>
<!--
Author: Tom Lane <tgl@sss.pgh.pa.us>
Branch: master [8f1911d5e] 2016-04-15 12:11:40 -0400
Branch: REL9_5_STABLE [8f8e65d34] 2016-04-15 12:11:27 -0400
Branch: REL9_4_STABLE [8eed31ffb] 2016-04-15 12:11:27 -0400
-->
<para>
Fix <command>ALTER TABLE ... REPLICA IDENTITY USING INDEX</> to
work properly if an index on OID is selected (David Rowley)
</para>
</listitem>
<listitem>
<!--
Author: Tom Lane <tgl@sss.pgh.pa.us>
Branch: master [93c301fc4] 2016-04-08 12:31:53 -0400
Branch: REL9_5_STABLE [0da7cf6e8] 2016-04-08 12:31:42 -0400
-->
<para>
Avoid possible misbehavior after failing to remove a tablespace symlink
(Tom Lane)
</para>
</listitem>
<listitem>
<!--
Author: Tom Lane <tgl@sss.pgh.pa.us>
Branch: master [6a3d3965d] 2016-04-14 19:42:21 -0400
Branch: REL9_5_STABLE [e7a456174] 2016-04-14 19:42:22 -0400
Branch: REL9_4_STABLE [00456911f] 2016-04-14 19:42:22 -0400
-->
<para>
Fix crash in logical decoding on alignment-picky platforms (Tom Lane,
Andres Freund)
</para>
<para>
The failure occurred only with a transaction large enough to spill to
disk and a primary-key change within that transaction.
</para>
</listitem>
<listitem>
<!--
Author: Andres Freund <andres@anarazel.de>
Branch: master [59455018a] 2016-04-28 22:11:18 -0700
Branch: REL9_5_STABLE [2e1b4adf3] 2016-04-28 22:11:18 -0700
Branch: REL9_4_STABLE [596f93605] 2016-04-28 22:11:18 -0700
-->
<para>
Avoid repeated requests for feedback from receiver while shutting down
walsender (Nick Cleaton)
</para>
</listitem>
<listitem>
<!--
Author: Tom Lane <tgl@sss.pgh.pa.us>
Branch: master [cbabb70f3] 2016-04-20 23:48:13 -0400
Branch: REL9_5_STABLE [0b8e0bf0a] 2016-04-20 23:48:13 -0400
Branch: REL9_4_STABLE [f05ac711b] 2016-04-20 23:48:13 -0400
Branch: REL9_3_STABLE [691073bd8] 2016-04-20 23:48:13 -0400
Branch: REL9_2_STABLE [1b22368ff] 2016-04-20 23:48:13 -0400
Branch: REL9_1_STABLE [4c1c9f80b] 2016-04-20 23:48:13 -0400
-->
<para>
Make <application>pg_regress</> use a startup timeout from the
<envar>PGCTLTIMEOUT</> environment variable, if that's set (Tom Lane)
</para>
<para>
This is for consistency with a behavior recently added
to <application>pg_ctl</>; it eases automated testing on slow machines.
</para>
</listitem>
<listitem>
<!--
Author: Tom Lane <tgl@sss.pgh.pa.us>
Branch: master [6cead413b] 2016-04-13 18:58:14 -0400
Branch: REL9_5_STABLE [fb0f63933] 2016-04-13 18:57:52 -0400
Branch: REL9_4_STABLE [5daf1012a] 2016-04-13 18:57:52 -0400
Branch: REL9_3_STABLE [34bf6bc56] 2016-04-13 18:57:52 -0400
Branch: REL9_2_STABLE [6bb42d520] 2016-04-13 18:57:52 -0400
Branch: REL9_1_STABLE [3ef1f3a3e] 2016-04-13 18:57:52 -0400
-->
<para>
Fix <application>pg_upgrade</> to correctly restore extension
membership for operator families containing only one operator class
(Tom Lane)
</para>
<para>
In such a case, the operator family was restored into the new database,
but it was no longer marked as part of the extension. This had no
immediate ill effects, but would cause later <application>pg_dump</>
runs to emit output that would cause (harmless) errors on restore.
</para>
</listitem>
<listitem>
<!--
Author: Tom Lane <tgl@sss.pgh.pa.us>
Branch: master [1a2c17f8e] 2016-05-06 22:05:56 -0400
Branch: REL9_5_STABLE [196870f2f] 2016-05-06 22:05:51 -0400
Branch: REL9_4_STABLE [e1aecebc0] 2016-05-06 22:05:51 -0400
Branch: REL9_3_STABLE [e1d88f983] 2016-05-06 22:05:51 -0400
-->
<para>
Fix <application>pg_upgrade</> to not fail when new-cluster TOAST rules
differ from old (Tom Lane)
</para>
<para>
<application>pg_upgrade</> had special-case code to handle the
situation where the new <productname>PostgreSQL</> version thinks that
a table should have a TOAST table while the old version did not. That
code was broken, so remove it, and instead do nothing in such cases;
there seems no reason to believe that we can't get along fine without
a TOAST table if that was okay according to the old version's rules.
</para>
</listitem>
<listitem>
<!--
Author: Noah Misch <noah@leadboat.com>
Branch: master [213c7df03] 2016-04-26 21:53:58 -0400
Branch: REL9_5_STABLE [f9989482d] 2016-04-26 21:54:07 -0400
-->
<para>
Fix atomic operations for PPC when using IBM's XLC compiler (Noah Misch)
</para>
</listitem>
<listitem>
<!--
Author: Tom Lane <tgl@sss.pgh.pa.us>
Branch: master [75c24d0f7] 2016-04-18 13:33:06 -0400
Branch: REL9_5_STABLE [c6a5b6677] 2016-04-18 13:33:07 -0400
Branch: REL9_4_STABLE [21b7f49eb] 2016-04-18 13:33:07 -0400
Branch: REL9_3_STABLE [6ec1ff852] 2016-04-18 13:33:07 -0400
Branch: REL9_2_STABLE [b24f7e280] 2016-04-18 13:33:07 -0400
-->
<para>
Reduce the number of SysV semaphores used by a build configured with
<option>--disable-spinlocks</> (Tom Lane)
</para>
</listitem>
<listitem>
<!--
Author: Tom Lane <tgl@sss.pgh.pa.us>
Branch: master [0ab3595e5] 2016-04-23 16:53:15 -0400
Branch: REL9_5_STABLE [bdf7744f7] 2016-04-23 16:53:15 -0400
Branch: REL9_4_STABLE [2a715371c] 2016-04-23 16:53:15 -0400
Branch: REL9_3_STABLE [252c35895] 2016-04-23 16:53:15 -0400
Branch: REL9_2_STABLE [0f5491283] 2016-04-23 16:53:15 -0400
Branch: REL9_1_STABLE [cbff4b708] 2016-04-23 16:53:15 -0400
-->
<para>
Rename internal function <function>strtoi()</>
to <function>strtoint()</> to avoid conflict with a NetBSD library
function (Thomas Munro)
</para>
</listitem>
<listitem>
<!--
Author: Tom Lane <tgl@sss.pgh.pa.us>
Branch: REL9_5_STABLE [20d4428f0] 2016-04-21 15:44:18 -0400
Branch: REL9_4_STABLE [5e2fb8862] 2016-04-21 15:44:18 -0400
Branch: REL9_3_STABLE [8f9518414] 2016-04-21 15:44:18 -0400
Branch: REL9_2_STABLE [6848827b4] 2016-04-21 15:44:18 -0400
Branch: REL9_1_STABLE [b2bb59bd5] 2016-04-21 15:44:18 -0400
Author: Tom Lane <tgl@sss.pgh.pa.us>
Branch: master [125ad539a] 2016-04-21 16:58:47 -0400
Branch: REL9_5_STABLE [d2e59cbc3] 2016-04-21 16:58:59 -0400
Branch: REL9_4_STABLE [56dee70d9] 2016-04-21 16:59:04 -0400
Branch: REL9_3_STABLE [82bf369ed] 2016-04-21 16:59:08 -0400
Branch: REL9_2_STABLE [b5ebc513d] 2016-04-21 16:59:13 -0400
Branch: REL9_1_STABLE [9028f404e] 2016-04-21 16:59:17 -0400
-->
<para>
Fix reporting of errors from <function>bind()</>
and <function>listen()</> system calls on Windows (Tom Lane)
</para>
</listitem>
<listitem>
<!--
Author: Andrew Dunstan <andrew@dunslane.net>
Branch: master [01a07e6c1] 2016-04-08 12:37:20 -0400
Branch: REL9_5_STABLE [849167943] 2016-04-08 12:36:59 -0400
Branch: REL9_4_STABLE [f79a7fa90] 2016-04-08 12:29:31 -0400
Branch: REL9_3_STABLE [ca5d6edbf] 2016-04-08 12:29:34 -0400
Branch: REL9_2_STABLE [9e77d0e3c] 2016-04-08 12:29:36 -0400
Branch: REL9_1_STABLE [d2189a802] 2016-04-08 12:33:29 -0400
-->
<para>
Reduce verbosity of compiler output when building with Microsoft Visual
Studio (Christian Ullrich)
</para>
</listitem>
<listitem>
<!--
Author: Andrew Dunstan <andrew@dunslane.net>
Branch: master [0fb54de9a] 2016-04-29 08:09:07 -0400
Branch: REL9_5_STABLE [da52474f3] 2016-04-29 08:06:25 -0400
Branch: master [7dc549238] 2016-04-29 09:49:31 -0400
Branch: REL9_5_STABLE [b71b4d852] 2016-04-29 09:50:38 -0400
Author: Tom Lane <tgl@sss.pgh.pa.us>
Branch: master [23b09e15b] 2016-04-28 11:50:58 -0400
Branch: REL9_5_STABLE [94a16531b] 2016-04-28 11:51:08 -0400
Branch: REL9_4_STABLE [65c2eeb00] 2016-04-28 11:51:13 -0400
Branch: REL9_3_STABLE [707c44fe2] 2016-04-28 11:51:17 -0400
Branch: REL9_2_STABLE [c563d97c6] 2016-04-28 11:51:22 -0400
Branch: REL9_1_STABLE [a6c6789c4] 2016-04-28 11:51:26 -0400
-->
<para>
Support building with Visual Studio 2015
(Michael Paquier, Petr Jel&iacute;nek)
</para>
<para>
Note that builds made with VS2015 will not run on Windows versions
before Windows Vista.
</para>
</listitem>
<listitem>
<!--
Author: Magnus Hagander <magnus@hagander.net>
Branch: master [9f633b404] 2016-04-22 05:18:59 -0400
Branch: REL9_5_STABLE [409c49c64] 2016-04-22 05:19:57 -0400
Branch: REL9_4_STABLE [c238a4101] 2016-04-22 05:20:07 -0400
Branch: REL9_3_STABLE [ab5c6d01f] 2016-04-22 05:20:18 -0400
-->
<para>
Fix <function>putenv()</> to work properly with Visual Studio 2013
(Michael Paquier)
</para>
</listitem>
<listitem>
<!--
Author: Tom Lane <tgl@sss.pgh.pa.us>
Branch: master [7abc15716] 2016-03-29 11:55:19 -0400
Branch: REL9_5_STABLE [491e84738] 2016-03-29 11:54:57 -0400
Branch: REL9_4_STABLE [2fed676c9] 2016-03-29 11:54:57 -0400
Branch: REL9_3_STABLE [11cc7bb88] 2016-03-29 11:54:57 -0400
Branch: REL9_2_STABLE [b4b06931e] 2016-03-29 11:54:58 -0400
Branch: REL9_1_STABLE [6cd30292b] 2016-03-29 11:54:58 -0400
-->
<para>
Avoid possibly-unsafe use of Windows' <function>FormatMessage()</>
function (Christian Ullrich)
</para>
<para>
Use the <literal>FORMAT_MESSAGE_IGNORE_INSERTS</> flag where
appropriate. No live bug is known to exist here, but it seems like a
good idea to be careful.
</para>
</listitem>
<listitem>
<!--
Author: Tom Lane <tgl@sss.pgh.pa.us>
Branch: master [98f158e41] 2016-05-05 20:08:58 -0400
Branch: REL9_5_STABLE [4edbb265c] 2016-05-05 20:09:10 -0400
Branch: REL9_4_STABLE [a5148e800] 2016-05-05 20:09:17 -0400
Branch: REL9_3_STABLE [d30c67af8] 2016-05-05 20:09:22 -0400
Branch: REL9_2_STABLE [29d154e36] 2016-05-05 20:09:27 -0400
Branch: REL9_1_STABLE [bfc39da64] 2016-05-05 20:09:32 -0400
-->
<para>
Update time zone data files to <application>tzdata</> release 2016d
for DST law changes in Russia and Venezuela. There are new zone
names <literal>Europe/Kirov</> and <literal>Asia/Tomsk</> to reflect
the fact that these regions now have different time zone histories from
adjacent regions.
</para>
</listitem>
</itemizedlist>
</sect2>
</sect1>
<sect1 id="release-9-5-2">
<title>Release 9.5.2</title>
@ -3401,7 +3902,7 @@ Add GUC and storage parameter to set the maximum size of GIN pending list.
<para>
<application>psql</> now fails if the file specified by
an <option>--output</> or <option>--log-file</> switch cannot be
written (Tom Lane, Daniel Verite)
written (Tom Lane, Daniel V&eacute;rit&eacute;)
</para>
<para>