<!-- doc/src/sgml/release-9.3.sgml -->

<!-- See header comment in release.sgml about typical markup -->

<sect1 id="release-9-3-24">

<title>Release 9.3.24</title>

<formalpara>

<title>Release date:</title>

<para>2018-08-09</para>

</formalpara>

<para>

This release contains a variety of fixes from 9.3.23.

For information about new features in the 9.3 major release, see

<xref linkend="release-9-3">.

</para>

<para>

The <productname>PostgreSQL</productname> community will stop releasing

updates for the 9.3.X release series shortly after September 2018.

Users are encouraged to update to a newer release branch soon.

</para>

<sect2>

<title>Migration to Version 9.3.24</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 9.3.23,

see <xref linkend="release-9-3-23">.

</para>

</sect2>

<sect2>

<title>Changes</title>

<itemizedlist>

<listitem>

<para>

Ensure that updates to the <structfield>relfrozenxid</structfield>

and <structfield>relminmxid</structfield> values

for <quote>nailed</quote> system catalogs are processed in a timely

fashion (Andres Freund)

</para>

<para>

Overoptimistic caching rules could prevent these updates from being

seen by other sessions, leading to spurious errors and/or data

corruption. The problem was significantly worse for shared catalogs,

such as <structname>pg_authid</structname>, because the stale cache

data could persist into new sessions as well as existing ones.

</para>

</listitem>

<listitem>

<para>

Fix case where a freshly-promoted standby crashes before having

completed its first post-recovery checkpoint (Michael Paquier, Kyotaro

Horiguchi, Pavan Deolasee, &Aacute;lvaro Herrera)

</para>

<para>

This led to a situation where the server did not think it had reached

a consistent database state during subsequent WAL replay, preventing

restart.

</para>

</listitem>

<listitem>

<para>

Avoid emitting a bogus WAL record when recycling an all-zero btree

page (Amit Kapila)

</para>

<para>

This mistake has been seen to cause assertion failures, and

potentially it could result in unnecessary query cancellations on hot

standby servers.

</para>

</listitem>

<listitem>

<para>

Improve performance of WAL replay for transactions that drop many

relations (Fujii Masao)

</para>

<para>

This change reduces the number of times that shared buffers are

scanned, so that it is of most benefit when that setting is large.

</para>

</listitem>

<listitem>

<para>

Improve performance of lock releasing in standby server WAL replay

(Thomas Munro)

</para>

</listitem>

<listitem>

<para>

Ensure a table's cached index list is correctly rebuilt after an index

creation fails partway through (Peter Geoghegan)

</para>

<para>

Previously, the failed index's OID could remain in the list, causing

problems later in the same session.

</para>

</listitem>

<listitem>

<para>

Fix misoptimization of equivalence classes involving composite-type

columns (Tom Lane)

</para>

<para>

This resulted in failure to recognize that an index on a composite

column could provide the sort order needed for a mergejoin on that

column.

</para>

</listitem>

<listitem>

<para>

Fix SQL-standard <literal>FETCH FIRST</literal> syntax to allow

parameters (<literal>$<replaceable>n</replaceable></literal>), as the

standard expects (Andrew Gierth)

</para>

</listitem>

<listitem>

<para>

Fix failure to schema-qualify some object names

in <function>getObjectDescription</function> output

(Kyotaro Horiguchi, Tom Lane)

</para>

<para>

Names of collations, conversions, and text search objects

were not schema-qualified when they should be.

</para>

</listitem>

<listitem>

<para>

Widen <command>COPY FROM</command>'s current-line-number counter

from 32 to 64 bits (David Rowley)

</para>

<para>

This avoids two problems with input exceeding 4G lines: <literal>COPY

FROM WITH HEADER</literal> would drop a line every 4G lines, not only

the first line, and error reports could show a wrong line number.

</para>

</listitem>

<listitem>

<para>

Add a string freeing function

to <application>ecpg</application>'s <filename>pgtypes</filename>

library, so that cross-module memory management problems can be

avoided on Windows (Takayuki Tsunakawa)

</para>

<para>

On Windows, crashes can ensue if the <function>free</function> call

for a given chunk of memory is not made from the same DLL

that <function>malloc</function>'ed the memory.

The <filename>pgtypes</filename> library sometimes returns strings

that it expects the caller to free, making it impossible to follow

this rule. Add a <function>PGTYPESchar_free()</function> function

that just wraps <function>free</function>, allowing applications

to follow this rule.

</para>

</listitem>

<listitem>

<para>

Fix <application>ecpg</application>'s support for <type>long

long</type> variables on Windows, as well as other platforms that

declare <function>strtoll</function>/<function>strtoull</function>

nonstandardly or not at all (Dang Minh Huong, Tom Lane)

</para>

</listitem>

<listitem>

<para>

Fix misidentification of SQL statement type in PL/pgSQL, when a rule

change causes a change in the semantics of a statement intra-session

(Tom Lane)

</para>

<para>

This error led to assertion failures, or in rare cases, failure to

enforce the <literal>INTO STRICT</literal> option as expected.

</para>

</listitem>

<listitem>

<para>

Fix password prompting in client programs so that echo is properly

disabled on Windows when <literal>stdin</literal> is not the

terminal (Matthew Stickney)

</para>

</listitem>

<listitem>

<para>

Further fix mis-quoting of values for list-valued GUC variables in

dumps (Tom Lane)

</para>

<para>

The previous fix for quoting of <varname>search_path</varname> and

other list-valued variables in <application>pg_dump</application>

output turned out to misbehave for empty-string list elements, and it

risked truncation of long file paths.

</para>

</listitem>

<listitem>

<para>

Make <application>pg_upgrade</application> check that the old server

was shut down cleanly (Bruce Momjian)

</para>

<para>

The previous check could be fooled by an immediate-mode shutdown.

</para>

</listitem>

<listitem>

<para>

Fix crash in <filename>contrib/ltree</filename>'s

<function>lca()</function> function when the input array is empty

(Pierre Ducroquet)

</para>

</listitem>

<listitem>

<para>

Fix various error-handling code paths in which an incorrect error code

might be reported (Michael Paquier, Tom Lane, Magnus Hagander)

</para>

</listitem>

<listitem>

<para>

Rearrange makefiles to ensure that programs link to freshly-built

libraries (such as <filename>libpq.so</filename>) rather than ones

that might exist in the system library directories (Tom Lane)

</para>

<para>

This avoids problems when building on platforms that supply old copies

of <productname>PostgreSQL</productname> libraries.

</para>

</listitem>

<listitem>

<para>

Update time zone data files to <application>tzdata</application>

release 2018e for DST law changes in North Korea, plus historical

corrections for Czechoslovakia.

</para>

<para>

This update includes a redefinition of <quote>daylight savings</quote>

in Ireland, as well as for some past years in Namibia and

Czechoslovakia. In those jurisdictions, legally standard time is

observed in summer, and daylight savings time in winter, so that the

daylight savings offset is one hour behind standard time not one hour

ahead. This does not affect either the actual UTC offset or the

timezone abbreviations in use; the only known effect is that

the <structfield>is_dst</structfield> column in

the <structname>pg_timezone_names</structname> view will now be true

in winter and false in summer in these cases.

</para>

</listitem>

</itemizedlist>

</sect2>

</sect1>

<sect1 id="release-9-3-23">

<title>Release 9.3.23</title>