<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><html xmlns="http://www.w3.org/1999/xhtml"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /><title>E.5. Release 11.8</title><link rel="stylesheet" type="text/css" href="stylesheet.css" /><link rev="made" href="pgsql-docs@lists.postgresql.org" /><meta name="generator" content="DocBook XSL Stylesheets Vsnapshot" /><link rel="prev" href="release-11-9.html" title="E.4. Release 11.9" /><link rel="next" href="release-11-7.html" title="E.6. Release 11.7" /></head><body><div xmlns="http://www.w3.org/TR/xhtml1/transitional" class="navheader"><table width="100%" summary="Navigation header"><tr><th colspan="5" align="center">E.5. Release 11.8</th></tr><tr><td width="10%" align="left"><a accesskey="p" href="release-11-9.html" title="E.4. Release 11.9">Prev</a> </td><td width="10%" align="left"><a accesskey="u" href="release.html" title="Appendix E. Release Notes">Up</a></td><th width="60%" align="center">Appendix E. Release Notes</th><td width="10%" align="right"><a accesskey="h" href="index.html" title="PostgreSQL 11.12 Documentation">Home</a></td><td width="10%" align="right"> <a accesskey="n" href="release-11-7.html" title="E.6. Release 11.7">Next</a></td></tr></table><hr></hr></div><div class="sect1" id="RELEASE-11-8"><div class="titlepage"><div><div><h2 class="title" style="clear: both">E.5. Release 11.8</h2></div></div></div><div class="toc"><dl class="toc"><dt><span class="sect2"><a href="release-11-8.html#id-1.11.6.9.4">E.5.1. Migration to Version 11.8</a></span></dt><dt><span class="sect2"><a href="release-11-8.html#id-1.11.6.9.5">E.5.2. Changes</a></span></dt></dl></div><p><strong>Release date: </strong>2020-05-14</p><p>
This release contains a variety of fixes from 11.7.
For information about new features in major release 11, see
<a class="xref" href="release-11.html" title="E.13. Release 11">Section E.13</a>.
</p><div class="sect2" id="id-1.11.6.9.4"><div class="titlepage"><div><div><h3 class="title">E.5.1. Migration to Version 11.8</h3></div></div></div><p>
A dump/restore is not required for those running 11.X.
</p><p>
However, if you are upgrading from a version earlier than 11.6,
see <a class="xref" href="release-11-6.html" title="E.7. Release 11.6">Section E.7</a>.
</p></div><div class="sect2" id="id-1.11.6.9.5"><div class="titlepage"><div><div><h3 class="title">E.5.2. Changes</h3></div></div></div><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem"><p>
Propagate <code class="command">ALTER TABLE ... SET STORAGE</code> to indexes
(Peter Eisentraut)
</p><p>
Non-expression index columns have always copied
the <code class="structfield">attstorage</code> property of their table
column at creation. Update them when <code class="command">ALTER TABLE ... SET
STORAGE</code> is done, to maintain consistency.
</p></li><li class="listitem"><p>
Preserve the <code class="structfield">indisclustered</code> setting of
indexes rewritten by <code class="command">ALTER TABLE</code> (Amit Langote,
Justin Pryzby)
</p><p>
Previously, <code class="command">ALTER TABLE</code> lost track of which index
had been used for <code class="command">CLUSTER</code>.
</p></li><li class="listitem"><p>
Preserve the replica identity properties of indexes rewritten
by <code class="command">ALTER TABLE</code> (Quan Zongliang, Peter Eisentraut)
</p></li><li class="listitem"><p>
Lock objects sooner during <code class="command">DROP OWNED BY</code>
(Álvaro Herrera)
</p><p>
This avoids failures in race-condition cases where another session is
deleting some of the same objects.
</p></li><li class="listitem"><p>
Fix error-case processing for <code class="command">CREATE ROLE ... IN
ROLE</code> (Andrew Gierth)
</p><p>
Some error cases would be reported as <span class="quote">“<span class="quote">unexpected node
type</span>”</span> or the like, instead of the intended message.
</p></li><li class="listitem"><p>
Ensure that when a partition is detached, any triggers cloned from
its formerly-parent table are removed (Justin Pryzby)
</p></li><li class="listitem"><p>
Ensure that unique indexes over partitioned tables match the
equality semantics of the partitioning key (Guancheng Luo)
</p><p>
This would only be an issue with index opclasses that have unusual
notions of equality, but it's wrong in theory, so check.
</p></li><li class="listitem"><p>
Ensure that members of the <code class="literal">pg_read_all_stats</code> role
can read all statistics views, as expected (Magnus Hagander)
</p><p>
The functions underlying
the <code class="structname">pg_stat_progress_*</code> views had not gotten
this memo.
</p></li><li class="listitem"><p>
Repair performance regression in
<code class="structname">information_schema</code>.<code class="structname">triggers</code>
view (Tom Lane)
</p><p>
This patch redefines that view so that an
outer <code class="literal">WHERE</code> clause constraining the table name
can be pushed down into the view, allowing its calculations to be
done only for triggers belonging to the table of interest rather
than all triggers in the database. In a database with many triggers
this would make a significant speed difference for queries of that
form. Since things worked that way before v11, this is a potential
performance regression. Users who find this to be a problem can fix
it by replacing the view definition (or, perhaps, just deleting and
reinstalling the whole <code class="structname">information_schema</code>
schema).
</p></li><li class="listitem"><p>
Fix full text search to handle NOT above a phrase search correctly
(Tom Lane)
</p><p>
Queries such as <code class="literal">!(foo<->bar)</code> failed to find
matching rows when implemented as a GiST or GIN index search.
</p></li><li class="listitem"><p>
Fix full text search for cases where a phrase search includes an
item with both prefix matching and a weight restriction (Tom Lane)
</p></li><li class="listitem"><p>
Fix <code class="function">ts_headline()</code> to make better headline
selections when working with phrase queries (Tom Lane)
</p></li><li class="listitem"><p>
Fix bugs in <code class="varname">gin_fuzzy_search_limit</code> processing
(Adé Heyward, Tom Lane)
</p><p>
A small value of <code class="varname">gin_fuzzy_search_limit</code> could
result in unexpected slowness due to unintentionally rescanning the
same index page many times. Another code path failed to apply the
intended filtering at all, possibly returning too many values.
</p></li><li class="listitem"><p>
Allow input of type <code class="type">circle</code> to accept the format
<span class="quote">“<span class="quote"><code class="literal">(<em class="replaceable"><code>x</code></em>,<em class="replaceable"><code>y</code></em>),<em class="replaceable"><code>r</code></em></code></span>”</span>
as the documentation says it does (David Zhang)
</p></li><li class="listitem"><p>
Make the <code class="function">get_bit()</code>
and <code class="function">set_bit()</code> functions cope
with <code class="type">bytea</code> strings longer than 256MB (Movead Li)
</p><p>
Since the bit number argument is only <code class="type">int4</code>, it's
impossible to use these functions to access bits beyond the first
256MB of a long <code class="type">bytea</code>. We'll widen the argument
to <code class="type">int8</code> in v13, but in the meantime, allow these
functions to work on the initial substring of a
long <code class="type">bytea</code>.
</p></li><li class="listitem"><p>
Ignore file-not-found errors in <code class="function">pg_ls_waldir()</code>
and allied functions (Tom Lane)
</p><p>
This prevents a race condition failure if a file is removed between
when we see its directory entry and when we attempt
to <code class="function">stat()</code> it.
</p></li><li class="listitem"><p>
Avoid possibly leaking an open-file descriptor for a directory
in <code class="function">pg_ls_dir()</code>,
<code class="function">pg_timezone_names()</code>,
<code class="function">pg_tablespace_databases()</code>, and allied functions
(Justin Pryzby)
</p></li><li class="listitem"><p>
Fix polymorphic-function type resolution to correctly infer the
actual type of an <code class="type">anyarray</code> output when given only
an <code class="type">anyrange</code> input (Tom Lane)
</p></li><li class="listitem"><p>
Avoid leakage of a hashed subplan's hash tables across multiple
executions (Andreas Karlsson, Tom Lane)
</p><p>
This mistake could result in severe memory bloat if a query
re-executed a hashed subplan enough times.
</p></li><li class="listitem"><p>
Avoid unlikely crash when <code class="command">REINDEX</code> is terminated
by a session-shutdown signal (Tom Lane)
</p></li><li class="listitem"><p>
Fix low-probability crash after constraint violation errors in
partitioned tables (Andres Freund)
</p></li><li class="listitem"><p>
Prevent printout of possibly-incorrect hash join table statistics
in <code class="command">EXPLAIN</code> (Konstantin Knizhnik, Tom Lane, Thomas
Munro)
</p></li><li class="listitem"><p>
Fix reporting of elapsed time for heap truncation steps
in <code class="command">VACUUM VERBOSE</code> (Tatsuhito Kasahara)
</p></li><li class="listitem"><p>
Fix possible undercounting of deleted B-tree index pages
in <code class="command">VACUUM VERBOSE</code> output (Peter Geoghegan)
</p></li><li class="listitem"><p>
Fix wrong bookkeeping for oldest deleted page in a B-tree index
(Peter Geoghegan)
</p><p>
This could cause subtly wrong decisions about
when <code class="command">VACUUM</code> can skip an index cleanup scan;
although it appears there may be no significant user-visible effects
from that.
</p></li><li class="listitem"><p>
Ensure that TimelineHistoryRead and TimelineHistoryWrite wait states
are reported in all code paths that read or write timeline history
files (Masahiro Ikeda)
</p></li><li class="listitem"><p>
Avoid possibly showing <span class="quote">“<span class="quote">waiting</span>”</span> twice in a process's
PS status (Masahiko Sawada)
</p></li><li class="listitem"><p>
Avoid failure if autovacuum tries to access a just-dropped temporary
schema (Tom Lane)
</p><p>
This hazard only arises if a superuser manually drops a temporary
schema; which isn't normal practice, but should work.
</p></li><li class="listitem"><p>
Avoid premature recycling of WAL segments during crash recovery
(Jehan-Guillaume de Rorthais)
</p><p>
WAL segments that become ready to be archived during crash recovery
were potentially recycled without being archived.
</p></li><li class="listitem"><p>
Avoid scanning irrelevant timelines during archive recovery (Kyotaro
Horiguchi)
</p><p>
This can eliminate many attempts to fetch non-existent WAL files from
archive storage, which is helpful if archive access is slow.
</p></li><li class="listitem"><p>
Remove bogus <span class="quote">“<span class="quote">subtransaction logged without previous top-level
txn record</span>”</span> error check in logical decoding (Arseny Sher,
Amit Kapila)
</p><p>
This condition is legitimately reachable in various scenarios, so
remove the check.
</p></li><li class="listitem"><p>
Ensure that a replication
slot's <code class="literal">io_in_progress_lock</code> is released in failure
code paths (Pavan Deolasee)
</p><p>
This could result in a walsender later becoming stuck waiting for
the lock.
</p></li><li class="listitem"><p>
Fix race conditions in synchronous standby management (Tom Lane)
</p><p>
During a change in the <code class="varname">synchronous_standby_names</code>
setting, there was a window in which wrong decisions could be made
about whether it is OK to release transactions that are waiting for
synchronous commit. Another hazard for similarly wrong decisions
existed if a walsender process exited and was immediately replaced
by another.
</p></li><li class="listitem"><p>
Ensure <code class="varname">nextXid</code> can't go backwards on a standby
server (Eka Palamadai)
</p><p>
This race condition could allow incorrect hot standby feedback
messages to be sent back to the primary server, potentially allowing
<code class="command">VACUUM</code> to run too soon on the primary.
</p></li><li class="listitem"><p>
Add missing SQLSTATE values to a few error reports (Sawada Masahiko)
</p></li><li class="listitem"><p>
Fix PL/pgSQL to reliably refuse to execute an event trigger function
as a plain function (Tom Lane)
</p></li><li class="listitem"><p>
Fix memory leak in <span class="application">libpq</span> when
using <code class="literal">sslmode=verify-full</code> (Roman Peshkurov)
</p><p>
Certificate verification during connection startup could leak some
memory. This would become an issue if a client process opened many
database connections during its lifetime.
</p></li><li class="listitem"><p>
Fix <span class="application">ecpg</span> to treat an argument of
just <span class="quote">“<span class="quote"><code class="literal">-</code></span>”</span> as meaning <span class="quote">“<span class="quote">read
from stdin</span>”</span> on all platforms (Tom Lane)
</p></li><li class="listitem"><p>
Allow tab-completion of the filename argument
to <span class="application">psql</span>'s <code class="command">\gx</code> command
(Vik Fearing)
</p></li><li class="listitem"><p>
Add <span class="application">pg_dump</span> support for <code class="command">ALTER
... DEPENDS ON EXTENSION</code> (Álvaro Herrera)
</p><p>
<span class="application">pg_dump</span> previously ignored dependencies added
this way, causing them to be forgotten during dump/restore or
<span class="application">pg_upgrade</span>.
</p></li><li class="listitem"><p>
Fix <span class="application">pg_dump</span> to dump comments on RLS
policy objects (Tom Lane)
</p></li><li class="listitem"><p>
In <span class="application">pg_dump</span>, postpone restore of event
triggers till the end (Fabrízio de Royes Mello, Hamid Akhtar,
Tom Lane)
</p><p>
This minimizes the risk that an event trigger could interfere with
the restoration of other objects.
</p></li><li class="listitem"><p>
Make <span class="application">pg_verify_checksums</span> skip tablespace
subdirectories that belong to a
different <span class="productname">PostgreSQL</span> major version
(Michael Banck, Bernd Helmle)
</p><p>
Such subdirectories don't really belong to our database cluster, and
so must not be processed.
</p></li><li class="listitem"><p>
Ignore temporary copies of <code class="filename">pg_internal.init</code>
in <span class="application">pg_verify_checksums</span> and related
programs (Michael Paquier)
</p></li><li class="listitem"><p>
Fix quoting of <code class="option">--encoding</code>, <code class="option">--lc-ctype</code>
and <code class="option">--lc-collate</code> values
in <span class="application">createdb</span> utility (Michael Paquier)
</p></li><li class="listitem"><p>
<code class="filename">contrib/lo</code>'s <code class="function">lo_manage()</code>
function crashed if called directly rather than as a trigger (Tom
Lane)
</p></li><li class="listitem"><p>
In <code class="filename">contrib/ltree</code>,
protect against overflow of <code class="type">ltree</code>
and <code class="type">lquery</code> length fields (Nikita Glukhov)
</p></li><li class="listitem"><p>
Work around failure in <code class="filename">contrib/pageinspect</code>'s
<code class="function">bt_metap()</code> function when an oldest_xact value
exceeds 2^31-1 (Peter Geoghegan)
</p><p>
Such XIDs will now be reported as negative integers, which isn't
great but it beats throwing an error. v13 will widen the output
argument to <code class="type">int8</code> to provide saner reporting.
</p></li><li class="listitem"><p>
Fix cache reference leak in <code class="filename">contrib/sepgsql</code>
(Michael Luo)
</p></li><li class="listitem"><p>
Avoid failures when dealing with Unix-style locale names on
Windows (Juan José Santamaría Flecha)
</p></li><li class="listitem"><p>
Use <span class="application">pkg-config</span>, if available, to
locate <span class="application">libxml2</span>
during <span class="application">configure</span> (Hugh McMaster, Tom
Lane, Peter Eisentraut)
</p><p>
If <span class="application">pkg-config</span> is not present or lacks
knowledge of <span class="application">libxml2</span>, we still
query <span class="application">xml2-config</span> as before.
</p><p>
This change could break build processes that try to
make <span class="productname">PostgreSQL</span> use a non-default version
of <span class="application">libxml2</span> by putting that
version's <span class="application">xml2-config</span> into
the <code class="varname">PATH</code>. Instead,
set <code class="varname">XML2_CONFIG</code> to point to the
non-default <span class="application">xml2-config</span>. That method
will work with either older or
newer <span class="productname">PostgreSQL</span> releases.
</p></li><li class="listitem"><p>
In MSVC builds, cope with spaces in the path name for Python
(Victor Wagner)
</p></li><li class="listitem"><p>
In MSVC builds, fix detection of Visual Studio version to work with
more language settings (Andrew Dunstan)
</p></li><li class="listitem"><p>
In MSVC builds, use <code class="literal">-Wno-deprecated</code> with bison
versions newer than 3.0, as non-Windows builds already do (Andrew
Dunstan)
</p></li><li class="listitem"><p>
Update time zone data files to <span class="application">tzdata</span>
release 2020a for DST law changes in Morocco and the Canadian Yukon,
plus historical corrections for Shanghai.
</p><p>
The America/Godthab zone has been renamed to America/Nuuk to reflect
current English usage; however, the old name remains available as a
compatibility link.
</p><p>
Also, update <span class="application">initdb</span>'s list of known
Windows time zone names to include recent additions, improving the
odds that it will correctly translate the system time zone setting
on that platform.
</p></li></ul></div></div></div><div xmlns="http://www.w3.org/TR/xhtml1/transitional" class="navfooter"><hr></hr><table width="100%" summary="Navigation footer"><tr><td width="40%" align="left"><a accesskey="p" href="release-11-9.html" title="E.4. Release 11.9">Prev</a> </td><td width="20%" align="center"><a accesskey="u" href="release.html" title="Appendix E. Release Notes">Up</a></td><td width="40%" align="right"> <a accesskey="n" href="release-11-7.html" title="E.6. Release 11.7">Next</a></td></tr><tr><td width="40%" align="left" valign="top">E.4. Release 11.9 </td><td width="20%" align="center"><a accesskey="h" href="index.html" title="PostgreSQL 11.12 Documentation">Home</a></td><td width="40%" align="right" valign="top"> E.6. Release 11.7</td></tr></table></div></body></html>
