Linux SRV-16 6.18.44-paas #1 SMP PREEMPT_DYNAMIC Thu Aug 13 10:08:12 UTC 2026 x86_64
/
proc
/
212
/
root
/
opt
/
postgresql-11.1-201812121535
/
doc
/
html
/
//proc/212/root/opt/postgresql-11.1-201812121535/doc/html/release-9-3-22.html
<?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.62. Release 9.3.22</title><link rel="stylesheet" type="text/css" href="stylesheet.css" /><link rev="made" href="pgsql-docs@postgresql.org" /><meta name="generator" content="DocBook XSL Stylesheets V1.79.1" /><link rel="prev" href="release-9-3-23.html" title="E.61. Release 9.3.23" /><link rel="next" href="release-9-3-21.html" title="E.63. Release 9.3.21" /></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.62. Release 9.3.22</th></tr><tr><td width="10%" align="left"><a accesskey="p" href="release-9-3-23.html" title="E.61. Release 9.3.23">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 11.1 Documentation">Home</a></td><td width="10%" align="right"> <a accesskey="n" href="release-9-3-21.html" title="E.63. Release 9.3.21">Next</a></td></tr></table><hr></hr></div><div class="sect1" id="RELEASE-9-3-22"><div class="titlepage"><div><div><h2 class="title" style="clear: both">E.62. Release 9.3.22</h2></div></div></div><div class="toc"><dl class="toc"><dt><span class="sect2"><a href="release-9-3-22.html#id-1.11.6.66.4">E.62.1. Migration to Version 9.3.22</a></span></dt><dt><span class="sect2"><a href="release-9-3-22.html#id-1.11.6.66.5">E.62.2. Changes</a></span></dt></dl></div><p><strong>Release date: </strong>2018-03-01</p><p> This release contains a variety of fixes from 9.3.21. For information about new features in the 9.3 major release, see <a class="xref" href="release-9-3.html" title="E.84. Release 9.3">Section E.84</a>. </p><div class="sect2" id="id-1.11.6.66.4"><div class="titlepage"><div><div><h3 class="title">E.62.1. Migration to Version 9.3.22</h3></div></div></div><p> A dump/restore is not required for those running 9.3.X. </p><p> However, if you run an installation in which not all users are mutually trusting, or if you maintain an application or extension that is intended for use in arbitrary situations, it is strongly recommended that you read the documentation changes described in the first changelog entry below, and take suitable steps to ensure that your installation or code is secure. </p><p> Also, the changes described in the second changelog entry below may cause functions used in index expressions or materialized views to fail during auto-analyze, or when reloading from a dump. After upgrading, monitor the server logs for such problems, and fix affected functions. </p><p> Also, if you are upgrading from a version earlier than 9.3.18, see <a class="xref" href="release-9-3-18.html" title="E.66. Release 9.3.18">Section E.66</a>. </p></div><div class="sect2" id="id-1.11.6.66.5"><div class="titlepage"><div><div><h3 class="title">E.62.2. Changes</h3></div></div></div><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem"><p> Document how to configure installations and applications to guard against search-path-dependent trojan-horse attacks from other users (Noah Misch) </p><p> Using a <code class="varname">search_path</code> setting that includes any schemas writable by a hostile user enables that user to capture control of queries and then run arbitrary SQL code with the permissions of the attacked user. While it is possible to write queries that are proof against such hijacking, it is notationally tedious, and it's very easy to overlook holes. Therefore, we now recommend configurations in which no untrusted schemas appear in one's search path. Relevant documentation appears in <a class="xref" href="ddl-schemas.html#DDL-SCHEMAS-PATTERNS" title="5.8.6. Usage Patterns">Section 5.8.6</a> (for database administrators and users), <a class="xref" href="libpq-connect.html" title="34.1. Database Connection Control Functions">Section 34.1</a> (for application authors), <a class="xref" href="extend-extensions.html#EXTEND-EXTENSIONS-STYLE" title="38.16.1. Defining Extension Objects">Section 38.16.1</a> (for extension authors), and <a class="xref" href="sql-createfunction.html" title="CREATE FUNCTION"><span class="refentrytitle">CREATE FUNCTION</span></a> (for authors of <code class="literal">SECURITY DEFINER</code> functions). (CVE-2018-1058) </p></li><li class="listitem"><p> Avoid use of insecure <code class="varname">search_path</code> settings in <span class="application">pg_dump</span> and other client programs (Noah Misch, Tom Lane) </p><p> <span class="application">pg_dump</span>, <span class="application">pg_upgrade</span>, <span class="application">vacuumdb</span> and other <span class="productname">PostgreSQL</span>-provided applications were themselves vulnerable to the type of hijacking described in the previous changelog entry; since these applications are commonly run by superusers, they present particularly attractive targets. To make them secure whether or not the installation as a whole has been secured, modify them to include only the <code class="structname">pg_catalog</code> schema in their <code class="varname">search_path</code> settings. Autovacuum worker processes now do the same, as well. </p><p> In cases where user-provided functions are indirectly executed by these programs — for example, user-provided functions in index expressions — the tighter <code class="varname">search_path</code> may result in errors, which will need to be corrected by adjusting those user-provided functions to not assume anything about what search path they are invoked under. That has always been good practice, but now it will be necessary for correct behavior. (CVE-2018-1058) </p></li><li class="listitem"><p> Fix misbehavior of concurrent-update rechecks with CTE references appearing in subplans (Tom Lane) </p><p> If a CTE (<code class="literal">WITH</code> clause reference) is used in an InitPlan or SubPlan, and the query requires a recheck due to trying to update or lock a concurrently-updated row, incorrect results could be obtained. </p></li><li class="listitem"><p> Fix planner failures with overlapping mergejoin clauses in an outer join (Tom Lane) </p><p> These mistakes led to <span class="quote">“<span class="quote">left and right pathkeys do not match in mergejoin</span>”</span> or <span class="quote">“<span class="quote">outer pathkeys do not match mergeclauses</span>”</span> planner errors in corner cases. </p></li><li class="listitem"><p> Repair <span class="application">pg_upgrade</span>'s failure to preserve <code class="structfield">relfrozenxid</code> for materialized views (Tom Lane, Andres Freund) </p><p> This oversight could lead to data corruption in materialized views after an upgrade, manifesting as <span class="quote">“<span class="quote">could not access status of transaction</span>”</span> or <span class="quote">“<span class="quote">found xmin from before relfrozenxid</span>”</span> errors. The problem would be more likely to occur in seldom-refreshed materialized views, or ones that were maintained only with <code class="command">REFRESH MATERIALIZED VIEW CONCURRENTLY</code>. </p><p> If such corruption is observed, it can be repaired by refreshing the materialized view (without <code class="literal">CONCURRENTLY</code>). </p></li><li class="listitem"><p> Fix incorrect reporting of PL/Python function names in error <code class="literal">CONTEXT</code> stacks (Tom Lane) </p><p> An error occurring within a nested PL/Python function call (that is, one reached via a SPI query from another PL/Python function) would result in a stack trace showing the inner function's name twice, rather than the expected results. Also, an error in a nested PL/Python <code class="literal">DO</code> block could result in a null pointer dereference crash on some platforms. </p></li><li class="listitem"><p> Allow <code class="filename">contrib/auto_explain</code>'s <code class="varname">log_min_duration</code> setting to range up to <code class="literal">INT_MAX</code>, or about 24 days instead of 35 minutes (Tom Lane) </p></li></ul></div></div></div><div class="navfooter"><hr /><table width="100%" summary="Navigation footer"><tr><td width="40%" align="left"><a accesskey="p" href="release-9-3-23.html">Prev</a> </td><td width="20%" align="center"><a accesskey="u" href="release.html">Up</a></td><td width="40%" align="right"> <a accesskey="n" href="release-9-3-21.html">Next</a></td></tr><tr><td width="40%" align="left" valign="top">E.61. Release 9.3.23 </td><td width="20%" align="center"><a accesskey="h" href="index.html">Home</a></td><td width="40%" align="right" valign="top"> E.63. Release 9.3.21</td></tr></table></div></body></html>