| Server IP : 172.67.134.114 / Your IP : 162.159.115.42 Web Server : Apache/2.4.37 System : Linux almalinux.duckdns.org 4.18.0-553.111.1.el8_10.x86_64 #1 SMP Sun Mar 8 20:06:07 EDT 2026 x86_64 User : ricodeal ( 1046) PHP Version : 7.4.33 Disable Function : NONE MySQL : OFF | cURL : ON | WGET : ON | Perl : ON | Python : ON | Sudo : ON | Pkexec : ON Directory : /usr/share/doc/postgresql-docs/html/ |
Upload File : |
<?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.16. Release 10.8</title><link rel="stylesheet" type="text/css" href="stylesheet.css" /><link rev="made" href="[email protected]" /><meta name="generator" content="DocBook XSL Stylesheets Vsnapshot" /><link rel="prev" href="release-10-9.html" title="E.15. Release 10.9" /><link rel="next" href="release-10-7.html" title="E.17. Release 10.7" /></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.16. Release 10.8</th></tr><tr><td width="10%" align="left"><a accesskey="p" href="release-10-9.html" title="E.15. Release 10.9">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 10.23 Documentation">Home</a></td><td width="10%" align="right"> <a accesskey="n" href="release-10-7.html" title="E.17. Release 10.7">Next</a></td></tr></table><hr></hr></div><div class="sect1" id="RELEASE-10-8"><div class="titlepage"><div><div><h2 class="title" style="clear: both">E.16. Release 10.8</h2></div></div></div><div class="toc"><dl class="toc"><dt><span class="sect2"><a href="release-10-8.html#id-1.11.6.20.4">E.16.1. Migration to Version 10.8</a></span></dt><dt><span class="sect2"><a href="release-10-8.html#id-1.11.6.20.5">E.16.2. Changes</a></span></dt></dl></div><p><strong>Release date: </strong>2019-05-09</p><p> This release contains a variety of fixes from 10.7.
For information about new features in major release 10, see
<a class="xref" href="release-10.html" title="E.24. Release 10">Section E.24</a>.
</p><div class="sect2" id="id-1.11.6.20.4"><div class="titlepage"><div><div><h3 class="title">E.16.1. Migration to Version 10.8</h3></div></div></div><p> A dump/restore is not required for those running 10.X.
</p><p> However, if you are upgrading from a version earlier than 10.6,
see <a class="xref" href="release-10-6.html" title="E.18. Release 10.6">Section E.18</a>.
</p></div><div class="sect2" id="id-1.11.6.20.5"><div class="titlepage"><div><div><h3 class="title">E.16.2. Changes</h3></div></div></div><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem"><p> Prevent row-level security policies from being bypassed via
selectivity estimators (Dean Rasheed)
</p><p> Some of the planner's selectivity estimators apply user-defined
operators to values found in <code class="structname">pg_statistic</code>
(e.g., most-common values). A leaky operator therefore can disclose
some of the entries in a data column, even if the calling user lacks
permission to read that column. In CVE-2017-7484 we added
restrictions to forestall that, but we failed to consider the
effects of row-level security. A user who has SQL permission to
read a column, but who is forbidden to see certain rows due to RLS
policy, might still learn something about those rows' contents via a
leaky operator. This patch further tightens the rules, allowing
leaky operators to be applied to statistics data only when there is
no relevant RLS policy. (CVE-2019-10130)
</p></li><li class="listitem"><p> Avoid catalog corruption when a temporary table with <code class="literal">ON
COMMIT DROP</code> and an identity column is created in a
single-statement transaction (Peter Eisentraut)
</p><p> This hazard was overlooked because the case is not actually useful,
since the temporary table would be dropped immediately after
creation.
</p></li><li class="listitem"><p> Avoid crash when an EPQ recheck is performed for a partitioned query
result relation (Amit Langote)
</p><p> This occurs when using <code class="literal">READ COMMITTED</code> isolation
level and another session has concurrently updated some of the
target row(s).
</p></li><li class="listitem"><p> Fix behavior for an <code class="command">UPDATE</code>
or <code class="command">DELETE</code> on an inheritance tree or partitioned
table in which every table can be excluded (Amit Langote, Tom Lane)
</p><p> In such cases, the query did not report the correct set of output
columns when a <code class="literal">RETURNING</code> clause was present, and
if there were any statement-level triggers that should be fired, it
didn't fire them.
</p></li><li class="listitem"><p> Avoid throwing incorrect errors for updates of temporary tables and
unlogged tables when a <code class="literal">FOR ALL TABLES</code> publication
exists (Peter Eisentraut)
</p><p> Such tables should be ignored for publication purposes, but some
parts of the code failed to do so.
</p></li><li class="listitem"><p> Fix handling of explicit <code class="literal">DEFAULT</code> items in
an <code class="command">INSERT ... VALUES</code> command with
multiple <code class="literal">VALUES</code> rows, if the target relation is
an updatable view (Amit Langote, Dean Rasheed)
</p><p> When the updatable view has no default for the column but its
underlying table has one, a single-row <code class="command">INSERT
... VALUES</code> will use the underlying table's default.
In the multi-row case, however, NULL was always used. Correct it to
act like the single-row case.
</p></li><li class="listitem"><p> Fix <code class="command">CREATE VIEW</code> to allow zero-column views
(Ashutosh Sharma)
</p><p> We should allow this for consistency with allowing zero-column
tables. Since a table can be converted to a view, zero-column views
could be created even with the restriction in place, leading to
dump/reload failures.
</p></li><li class="listitem"><p> Add missing support for <code class="command">CREATE TABLE IF NOT EXISTS ... AS
EXECUTE ...</code> (Andreas Karlsson)
</p><p> The combination of <code class="literal">IF NOT EXISTS</code>
and <code class="literal">EXECUTE</code> should work, but the grammar omitted
it.
</p></li><li class="listitem"><p> Ensure that sub-<code class="command">SELECT</code>s appearing in
row-level-security policy expressions are executed with the correct
user's permissions (Dean Rasheed)
</p><p> Previously, if the table having the RLS policy was accessed via a
view, such checks might be executed as the user calling the view,
not as the view owner as they should be.
</p></li><li class="listitem"><p> Accept XML documents as valid values of type <code class="type">xml</code>
when <code class="varname">xmloption</code> is set
to <code class="literal">content</code>, as required by SQL:2006 and later
(Chapman Flack)
</p><p> Previously <span class="productname">PostgreSQL</span> followed the
SQL:2003 definition, which doesn't allow this. But that creates a
serious problem for dump/restore: there is no setting
of <code class="varname">xmloption</code> that will accept all valid XML data.
Hence, switch to the 2006 definition.
</p><p> <span class="application">pg_dump</span> is also modified to emit
<code class="literal">SET xmloption = content</code> while restoring data,
ensuring that dump/restore works even if the prevailing
setting is <code class="literal">document</code>.
</p></li><li class="listitem"><p> Improve server's startup-time checks for whether a pre-existing
shared memory segment is still in use (Noah Misch)
</p><p> The postmaster is now more likely to detect that there are still
active processes from a previous postmaster incarnation, even if
the <code class="filename">postmaster.pid</code> file has been removed.
</p></li><li class="listitem"><p> Avoid counting parallel workers' transactions as separate
transactions (Haribabu Kommi)
</p></li><li class="listitem"><p> Fix incompatibility of GIN-index WAL records (Alexander Korotkov)
</p><p> A fix applied in February's minor releases was not sufficiently
careful about backwards compatibility, leading to problems if a
standby server of that vintage reads GIN page-deletion WAL records
generated by a primary server of a previous minor release.
</p></li><li class="listitem"><p> Fix possible crash while executing a <code class="command">SHOW</code> command
in a replication connection (Michael Paquier)
</p></li><li class="listitem"><p> Avoid memory leak when a partition's relation cache entry is rebuilt
(Amit Langote, Tom Lane)
</p></li><li class="listitem"><p> Tolerate <code class="literal">EINVAL</code> and <code class="literal">ENOSYS</code>
error results, where appropriate, for <code class="function">fsync</code>
and <code class="function">sync_file_range</code> calls
(Thomas Munro, James Sewell)
</p><p> The previous change to panic on file synchronization failures turns
out to have been excessively paranoid for certain cases where a
failure is predictable and essentially means <span class="quote">“<span class="quote">operation not
supported</span>”</span>.
</p></li><li class="listitem"><p> Report correct relation name in
autovacuum's <code class="structname">pg_stat_activity</code> display
during BRIN summarize operations (Álvaro Herrera)
</p></li><li class="listitem"><p> Fix <span class="quote">“<span class="quote">failed to build any <em class="replaceable"><code>N</code></em>-way
joins</span>”</span> planner failures with lateral references leading out
of <code class="literal">FULL</code> outer joins (Tom Lane)
</p></li><li class="listitem"><p> Fix misplanning of queries in which a set-returning function is
applied to a relation that is provably empty (Tom Lane, Julien
Rouhaud)
</p><p> In v10, this oversight only led to slightly inefficient plans, but
in v11 it could cause <span class="quote">“<span class="quote">set-valued function called in context
that cannot accept a set</span>”</span> errors.
</p></li><li class="listitem"><p> Check the appropriate user's permissions when enforcing rules about
letting a leaky operator see <code class="structname">pg_statistic</code>
data (Dean Rasheed)
</p><p> When an underlying table is being accessed via a view, consider the
privileges of the view owner while deciding whether leaky operators
may be applied to the table's statistics data, rather than the
privileges of the user making the query. This makes the planner's
rules about what data is visible match up with the executor's,
avoiding unnecessarily-poor plans.
</p></li><li class="listitem"><p> Speed up planning when there are many equality conditions and many
potentially-relevant foreign key constraints (David Rowley)
</p></li><li class="listitem"><p> Avoid O(N^2) performance issue when rolling back a transaction that
created many tables (Tomas Vondra)
</p></li><li class="listitem"><p> Fix corner-case server crashes in dynamic shared memory
allocation (Thomas Munro, Robert Haas)
</p></li><li class="listitem"><p> Fix race conditions in management of dynamic shared memory
(Thomas Munro)
</p><p> These could lead to <span class="quote">“<span class="quote">dsa_area could not attach to
segment</span>”</span> or <span class="quote">“<span class="quote">cannot unpin a segment that is not
pinned</span>”</span> errors.
</p></li><li class="listitem"><p> Fix race condition in which a hot-standby postmaster could fail to
shut down after receiving a smart-shutdown request (Tom Lane)
</p></li><li class="listitem"><p> Fix possible crash
when <code class="function">pg_identify_object_as_address()</code> is given
invalid input (Álvaro Herrera)
</p></li><li class="listitem"><p> Fix possible <span class="quote">“<span class="quote">could not access status of transaction</span>”</span>
failures in <code class="function">txid_status()</code> (Thomas Munro)
</p></li><li class="listitem"><p> Tighten validation of encoded SCRAM-SHA-256 and MD5 passwords
(Jonathan Katz)
</p><p> A password string that had the right initial characters could be
mistaken for one that is correctly hashed into SCRAM-SHA-256 or MD5
format. The password would be accepted but would be unusable later.
</p></li><li class="listitem"><p> Fix handling of <code class="varname">lc_time</code> settings that imply an
encoding different from the database's encoding (Juan José
Santamaría Flecha, Tom Lane)
</p><p> Localized month or day names that include non-ASCII characters
previously caused unexpected errors or wrong output in such locales.
</p></li><li class="listitem"><p> Fix incorrect <code class="varname">operator_precedence_warning</code> checks
involving unary minus operators (Rikard Falkeborn)
</p></li><li class="listitem"><p> Disallow <code class="literal">NaN</code> as a value for floating-point server
parameters (Tom Lane)
</p></li><li class="listitem"><p> Rearrange <code class="command">REINDEX</code> processing to avoid assertion
failures when reindexing individual indexes
of <code class="structname">pg_class</code> (Andres Freund, Tom Lane)
</p></li><li class="listitem"><p> Fix planner assertion failure for parameterized dummy paths (Tom Lane)
</p></li><li class="listitem"><p> Insert correct test function in the result
of <code class="function">SnapBuildInitialSnapshot()</code> (Antonin Houska)
</p><p> No core code cares about this, but some extensions do.
</p></li><li class="listitem"><p> Fix intermittent <span class="quote">“<span class="quote">could not reattach to shared memory</span>”</span>
session startup failures on Windows (Noah Misch)
</p><p> A previously unrecognized source of these failures is creation of
thread stacks for a process's default thread pool. Arrange for such
stacks to be allocated in a different memory region.
</p></li><li class="listitem"><p> Fix error detection in directory scanning on Windows (Konstantin
Knizhnik)
</p><p> Errors, such as lack of permissions to read the directory, were not
detected or reported correctly; instead the code silently acted as
though the directory were empty.
</p></li><li class="listitem"><p> Fix grammar problems in <span class="application">ecpg</span> (Tom Lane)
</p><p> A missing semicolon led to mistranslation
of <code class="literal">SET <em class="replaceable"><code>variable</code></em> =
DEFAULT</code> (but
not <code class="literal">SET <em class="replaceable"><code>variable</code></em> TO
DEFAULT</code>) in <span class="application">ecpg</span> programs,
producing syntactically invalid output that the server would reject.
Additionally, in a <code class="command">DROP TYPE</code> or <code class="command">DROP
DOMAIN</code> command that listed multiple type names, only the
first type name was actually processed.
</p></li><li class="listitem"><p> Sync <span class="application">ecpg</span>'s syntax for <code class="command">CREATE
TABLE AS</code> with the server's (Daisuke Higuchi)
</p></li><li class="listitem"><p> Fix possible buffer overruns in <span class="application">ecpg</span>'s
processing of include filenames (Liu Huailing, Fei Wu)
</p></li><li class="listitem"><p> Avoid crash in <code class="filename">contrib/postgres_fdw</code> when a
query using remote grouping or aggregation has
a <code class="literal">SELECT</code>-list item that is an uncorrelated
sub-select, outer reference, or parameter symbol (Tom Lane)
</p></li><li class="listitem"><p> Avoid crash in <code class="filename">contrib/vacuumlo</code> if
an <code class="function">lo_unlink()</code> call failed (Tom Lane)
</p></li><li class="listitem"><p> Sync our copy of the timezone library with IANA tzcode release 2019a
(Tom Lane)
</p><p> This corrects a small bug in <span class="application">zic</span> that
caused it to output an incorrect year-2440 transition in
the <code class="literal">Africa/Casablanca</code> zone, and adds support
for <span class="application">zic</span>'s new <code class="option">-r</code> option.
</p></li><li class="listitem"><p> Update time zone data files to <span class="application">tzdata</span>
release 2019a for DST law changes in Palestine and Metlakatla,
plus historical corrections for Israel.
</p><p> <code class="literal">Etc/UCT</code> is now a backward-compatibility link
to <code class="literal">Etc/UTC</code>, instead of being a separate zone that
generates the abbreviation <code class="literal">UCT</code>, which nowadays is
typically a typo. <span class="productname">PostgreSQL</span> will still
accept <code class="literal">UCT</code> as an input zone abbreviation, but it
won't output it.
</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-10-9.html" title="E.15. Release 10.9">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-10-7.html" title="E.17. Release 10.7">Next</a></td></tr><tr><td width="40%" align="left" valign="top">E.15. Release 10.9 </td><td width="20%" align="center"><a accesskey="h" href="index.html" title="PostgreSQL 10.23 Documentation">Home</a></td><td width="40%" align="right" valign="top"> E.17. Release 10.7</td></tr></table></div></body></html>