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-9-5-6.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.31. Release 9.5.6</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-9-5-7.html" title="E.30. Release 9.5.7" /><link rel="next" href="release-9-5-5.html" title="E.32. Release 9.5.5" /></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.31. Release 9.5.6</th></tr><tr><td width="10%" align="left"><a accesskey="p" href="release-9-5-7.html" title="E.30. Release 9.5.7">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-9-5-5.html" title="E.32. Release 9.5.5">Next</a></td></tr></table><hr></hr></div><div class="sect1" id="RELEASE-9-5-6"><div class="titlepage"><div><div><h2 class="title" style="clear: both">E.31. Release 9.5.6</h2></div></div></div><div class="toc"><dl class="toc"><dt><span class="sect2"><a href="release-9-5-6.html#id-1.11.6.35.4">E.31.1. Migration to Version 9.5.6</a></span></dt><dt><span class="sect2"><a href="release-9-5-6.html#id-1.11.6.35.5">E.31.2. Changes</a></span></dt></dl></div><p><strong>Release date: </strong>2017-02-09</p><p>
   This release contains a variety of fixes from 9.5.5.
   For information about new features in the 9.5 major release, see
   <a class="xref" href="release-9-5.html" title="E.37. Release 9.5">Section E.37</a>.
  </p><div class="sect2" id="id-1.11.6.35.4"><div class="titlepage"><div><div><h3 class="title">E.31.1. Migration to Version 9.5.6</h3></div></div></div><p>
    A dump/restore is not required for those running 9.5.X.
   </p><p>
    However, if your installation has been affected by the bug described in
    the first changelog entry below, then after updating you may need
    to take action to repair corrupted indexes.
   </p><p>
    Also, if you are upgrading from a version earlier than 9.5.5,
    see <a class="xref" href="release-9-5-5.html" title="E.32. Release 9.5.5">Section E.32</a>.
   </p></div><div class="sect2" id="id-1.11.6.35.5"><div class="titlepage"><div><div><h3 class="title">E.31.2. Changes</h3></div></div></div><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem"><p>
      Fix a race condition that could cause indexes built
      with <code class="command">CREATE INDEX CONCURRENTLY</code> to be corrupt
      (Pavan Deolasee, Tom Lane)
     </p><p>
      If <code class="command">CREATE INDEX CONCURRENTLY</code> was used to build an index
      that depends on a column not previously indexed, then rows
      updated by transactions that ran concurrently with
      the <code class="command">CREATE INDEX</code> command could have received incorrect
      index entries.  If you suspect this may have happened, the most
      reliable solution is to rebuild affected indexes after installing
      this update.
     </p></li><li class="listitem"><p>
      Ensure that the special snapshot used for catalog scans is not
      invalidated by premature data pruning (Tom Lane)
     </p><p>
      Backends failed to account for this snapshot when advertising their
      oldest xmin, potentially allowing concurrent vacuuming operations to
      remove data that was still needed.  This led to transient failures
      along the lines of <span class="quote">“<span class="quote">cache lookup failed for relation 1255</span>”</span>.
     </p></li><li class="listitem"><p>
      Fix incorrect WAL logging for BRIN indexes (Kuntal Ghosh)
     </p><p>
      The WAL record emitted for a BRIN <span class="quote">“<span class="quote">revmap</span>”</span> page when moving an
      index tuple to a different page was incorrect.  Replay would make the
      related portion of the index useless, forcing it to be recomputed.
     </p></li><li class="listitem"><p>
      Unconditionally WAL-log creation of the <span class="quote">“<span class="quote">init fork</span>”</span> for an
      unlogged table (Michael Paquier)
     </p><p>
      Previously, this was skipped when <a class="xref" href="runtime-config-wal.html#GUC-WAL-LEVEL">wal_level</a>
      = <code class="literal">minimal</code>, but actually it's necessary even in that case
      to ensure that the unlogged table is properly reset to empty after a
      crash.
     </p></li><li class="listitem"><p>
      Reduce interlocking on standby servers during the replay of btree
      index vacuuming operations (Simon Riggs)
     </p><p>
      This change avoids substantial replication delays that sometimes
      occurred while replaying such operations.
     </p></li><li class="listitem"><p>
      If the stats collector dies during hot standby, restart it (Takayuki
      Tsunakawa)
     </p></li><li class="listitem"><p>
      Ensure that hot standby feedback works correctly when it's enabled at
      standby server start (Ants Aasma, Craig Ringer)
     </p></li><li class="listitem"><p>
      Check for interrupts while hot standby is waiting for a conflicting
      query (Simon Riggs)
     </p></li><li class="listitem"><p>
      Avoid constantly respawning the autovacuum launcher in a corner case
      (Amit Khandekar)
     </p><p>
      This fix avoids problems when autovacuum is nominally off and there
      are some tables that require freezing, but all such tables are
      already being processed by autovacuum workers.
     </p></li><li class="listitem"><p>
      Fix check for when an extension member object can be dropped (Tom Lane)
     </p><p>
      Extension upgrade scripts should be able to drop member objects,
      but this was disallowed for serial-column sequences, and possibly
      other cases.
     </p></li><li class="listitem"><p>
      Make sure <code class="command">ALTER TABLE</code> preserves index tablespace
      assignments when rebuilding indexes (Tom Lane, Michael Paquier)
     </p><p>
      Previously, non-default settings
      of <a class="xref" href="runtime-config-client.html#GUC-DEFAULT-TABLESPACE">default_tablespace</a> could result in broken
      indexes.
     </p></li><li class="listitem"><p>
      Fix incorrect updating of trigger function properties when changing a
      foreign-key constraint's deferrability properties with <code class="command">ALTER
      TABLE ... ALTER CONSTRAINT</code> (Tom Lane)
     </p><p>
      This led to odd failures during subsequent exercise of the foreign
      key, as the triggers were fired at the wrong times.
     </p></li><li class="listitem"><p>
      Prevent dropping a foreign-key constraint if there are pending
      trigger events for the referenced relation (Tom Lane)
     </p><p>
      This avoids <span class="quote">“<span class="quote">could not find trigger <em class="replaceable"><code>NNN</code></em></span>”</span>
      or <span class="quote">“<span class="quote">relation <em class="replaceable"><code>NNN</code></em> has no triggers</span>”</span> errors.
     </p></li><li class="listitem"><p>
      Fix <code class="command">ALTER TABLE ... SET DATA TYPE ... USING</code> when child
      table has different column ordering than the parent
      (Álvaro Herrera)
     </p><p>
      Failure to adjust the column numbering in the <code class="literal">USING</code>
      expression led to errors,
      typically <span class="quote">“<span class="quote">attribute <em class="replaceable"><code>N</code></em> has wrong type</span>”</span>.
     </p></li><li class="listitem"><p>
      Fix processing of OID column when a table with OIDs is associated to
      a parent with OIDs via <code class="command">ALTER TABLE ... INHERIT</code> (Amit
      Langote)
     </p><p>
      The OID column should be treated the same as regular user columns in
      this case, but it wasn't, leading to odd behavior in later
      inheritance changes.
     </p></li><li class="listitem"><p>
      Fix <code class="command">CREATE OR REPLACE VIEW</code> to update the view query
      before attempting to apply the new view options (Dean Rasheed)
     </p><p>
      Previously the command would fail if the new options were
      inconsistent with the old view definition.
     </p></li><li class="listitem"><p>
      Report correct object identity during <code class="command">ALTER TEXT SEARCH
      CONFIGURATION</code> (Artur Zakirov)
     </p><p>
      The wrong catalog OID was reported to extensions such as logical
      decoding.
     </p></li><li class="listitem"><p>
      Fix commit timestamp mechanism to not fail when queried about
      the special XIDs <code class="literal">FrozenTransactionId</code>
      and <code class="literal">BootstrapTransactionId</code> (Craig Ringer)
     </p></li><li class="listitem"><p>
      Check for serializability conflicts before reporting
      constraint-violation failures (Thomas Munro)
     </p><p>
      When using serializable transaction isolation, it is desirable
      that any error due to concurrent transactions should manifest
      as a serialization failure, thereby cueing the application that
      a retry might succeed.  Unfortunately, this does not reliably
      happen for duplicate-key failures caused by concurrent insertions.
      This change ensures that such an error will be reported as a
      serialization error if the application explicitly checked for
      the presence of a conflicting key (and did not find it) earlier
      in the transaction.
     </p></li><li class="listitem"><p>
      Fix incorrect use of view reloptions as regular table reloptions (Tom
      Lane)
     </p><p>
      The symptom was spurious <span class="quote">“<span class="quote">ON CONFLICT is not supported on table
      ... used as a catalog table</span>”</span> errors when the target
      of <code class="command">INSERT ... ON CONFLICT</code> is a view with cascade option.
     </p></li><li class="listitem"><p>
      Fix incorrect <span class="quote">“<span class="quote">target lists can have at most <em class="replaceable"><code>N</code></em>
      entries</span>”</span> complaint when using <code class="literal">ON CONFLICT</code> with
      wide tables (Tom Lane)
     </p></li><li class="listitem"><p>
      Prevent multicolumn expansion of <em class="replaceable"><code>foo</code></em><code class="literal">.*</code> in
      an <code class="command">UPDATE</code> source expression (Tom Lane)
     </p><p>
      This led to <span class="quote">“<span class="quote">UPDATE target count mismatch --- internal
      error</span>”</span>.  Now the syntax is understood as a whole-row variable,
      as it would be in other contexts.
     </p></li><li class="listitem"><p>
      Ensure that column typmods are determined accurately for
      multi-row <code class="literal">VALUES</code> constructs (Tom Lane)
     </p><p>
      This fixes problems occurring when the first value in a column has a
      determinable typmod (e.g., length for a <code class="type">varchar</code> value) but
      later values don't share the same limit.
     </p></li><li class="listitem"><p>
      Throw error for an unfinished Unicode surrogate pair at the end of a
      Unicode string (Tom Lane)
     </p><p>
      Normally, a Unicode surrogate leading character must be followed by a
      Unicode surrogate trailing character, but the check for this was
      missed if the leading character was the last character in a Unicode
      string literal (<code class="literal">U&amp;'...'</code>) or Unicode identifier
      (<code class="literal">U&amp;"..."</code>).
     </p></li><li class="listitem"><p>
      Ensure that a purely negative text search query, such
      as <code class="literal">!foo</code>, matches empty <code class="type">tsvector</code>s (Tom Dunstan)
     </p><p>
      Such matches were found by GIN index searches, but not by sequential
      scans or GiST index searches.
     </p></li><li class="listitem"><p>
      Prevent crash when <code class="function">ts_rewrite()</code> replaces a non-top-level
      subtree with an empty query (Artur Zakirov)
     </p></li><li class="listitem"><p>
      Fix performance problems in <code class="function">ts_rewrite()</code> (Tom Lane)
     </p></li><li class="listitem"><p>
      Fix <code class="function">ts_rewrite()</code>'s handling of nested NOT operators
      (Tom Lane)
     </p></li><li class="listitem"><p>
      Improve speed of user-defined aggregates that
      use <code class="function">array_append()</code> as transition function (Tom Lane)
     </p></li><li class="listitem"><p>
      Fix <code class="function">array_fill()</code> to handle empty arrays properly (Tom Lane)
     </p></li><li class="listitem"><p>
      Fix possible crash in <code class="function">array_position()</code>
      or <code class="function">array_positions()</code> when processing arrays of records
      (Junseok Yang)
     </p></li><li class="listitem"><p>
      Fix one-byte buffer overrun in <code class="function">quote_literal_cstr()</code>
      (Heikki Linnakangas)
     </p><p>
      The overrun occurred only if the input consisted entirely of single
      quotes and/or backslashes.
     </p></li><li class="listitem"><p>
      Prevent multiple calls of <code class="function">pg_start_backup()</code>
      and <code class="function">pg_stop_backup()</code> from running concurrently (Michael
      Paquier)
     </p><p>
      This avoids an assertion failure, and possibly worse things, if
      someone tries to run these functions in parallel.
     </p></li><li class="listitem"><p>
      Disable transform that attempted to remove no-op <code class="literal">AT TIME
      ZONE</code> conversions (Tom Lane)
     </p><p>
      This resulted in wrong answers when the simplified expression was
      used in an index condition.
     </p></li><li class="listitem"><p>
      Avoid discarding <code class="type">interval</code>-to-<code class="type">interval</code> casts
      that aren't really no-ops (Tom Lane)
     </p><p>
      In some cases, a cast that should result in zeroing out
      low-order <code class="type">interval</code> fields was mistakenly deemed to be a
      no-op and discarded.  An example is that casting from <code class="type">INTERVAL
      MONTH</code> to <code class="type">INTERVAL YEAR</code> failed to clear the months field.
     </p></li><li class="listitem"><p>
      Fix bugs in transmitting GUC parameter values to parallel workers
      (Michael Paquier, Tom Lane)
     </p></li><li class="listitem"><p>
      Ensure that cached plans are invalidated by changes in foreign-table
      options (Amit Langote, Etsuro Fujita, Ashutosh Bapat)
     </p></li><li class="listitem"><p>
      Fix <span class="application">pg_dump</span> to dump user-defined casts and transforms
      that use built-in functions (Stephen Frost)
     </p></li><li class="listitem"><p>
      Fix <span class="application">pg_restore</span> with <code class="option">--create --if-exists</code>
      to behave more sanely if an archive contains
      unrecognized <code class="command">DROP</code> commands (Tom Lane)
     </p><p>
      This doesn't fix any live bug, but it may improve the behavior in
      future if <span class="application">pg_restore</span> is used with an archive
      generated by a later <span class="application">pg_dump</span> version.
     </p></li><li class="listitem"><p>
      Fix <span class="application">pg_basebackup</span>'s rate limiting in the presence of
      slow I/O (Antonin Houska)
     </p><p>
      If disk I/O was transiently much slower than the specified rate
      limit, the calculation overflowed, effectively disabling the rate
      limit for the rest of the run.
     </p></li><li class="listitem"><p>
      Fix <span class="application">pg_basebackup</span>'s handling of
      symlinked <code class="filename">pg_stat_tmp</code> and <code class="filename">pg_replslot</code>
      subdirectories (Magnus Hagander, Michael Paquier)
     </p></li><li class="listitem"><p>
      Fix possible <span class="application">pg_basebackup</span> failure on standby
      server when including WAL files (Amit Kapila, Robert Haas)
     </p></li><li class="listitem"><p>
      Fix possible mishandling of expanded arrays in domain check
      constraints and <code class="literal">CASE</code> execution (Tom Lane)
     </p><p>
      It was possible for a PL/pgSQL function invoked in these contexts to
      modify or even delete an array value that needs to be preserved for
      additional operations.
     </p></li><li class="listitem"><p>
      Fix nested uses of PL/pgSQL functions in contexts such as domain
      check constraints evaluated during assignment to a PL/pgSQL variable
      (Tom Lane)
     </p></li><li class="listitem"><p>
      Ensure that the Python exception objects we create for PL/Python are
      properly reference-counted (Rafa de la Torre, Tom Lane)
     </p><p>
      This avoids failures if the objects are used after a Python garbage
      collection cycle has occurred.
     </p></li><li class="listitem"><p>
      Fix PL/Tcl to support triggers on tables that have <code class="literal">.tupno</code>
      as a column name (Tom Lane)
     </p><p>
      This matches the (previously undocumented) behavior of
      PL/Tcl's <code class="command">spi_exec</code> and <code class="command">spi_execp</code> commands,
      namely that a magic <code class="literal">.tupno</code> column is inserted only if
      there isn't a real column named that.
     </p></li><li class="listitem"><p>
      Allow DOS-style line endings in <code class="filename">~/.pgpass</code> files,
      even on Unix (Vik Fearing)
     </p><p>
      This change simplifies use of the same password file across Unix and
      Windows machines.
     </p></li><li class="listitem"><p>
      Fix one-byte buffer overrun if <span class="application">ecpg</span> is given a file
      name that ends with a dot (Takayuki Tsunakawa)
     </p></li><li class="listitem"><p>
      Fix <span class="application">psql</span>'s tab completion for <code class="command">ALTER DEFAULT
      PRIVILEGES</code> (Gilles Darold, Stephen Frost)
     </p></li><li class="listitem"><p>
      In <span class="application">psql</span>, treat an empty or all-blank setting of
      the <code class="envar">PAGER</code> environment variable as meaning <span class="quote">“<span class="quote">no
      pager</span>”</span> (Tom Lane)
     </p><p>
      Previously, such a setting caused output intended for the pager to
      vanish entirely.
     </p></li><li class="listitem"><p>
      Improve <code class="filename">contrib/dblink</code>'s reporting of
      low-level <span class="application">libpq</span> errors, such as out-of-memory
      (Joe Conway)
     </p></li><li class="listitem"><p>
      Teach <code class="filename">contrib/dblink</code> to ignore irrelevant server options
      when it uses a <code class="filename">contrib/postgres_fdw</code> foreign server as
      the source of connection options (Corey Huinker)
     </p><p>
      Previously, if the foreign server object had options that were not
      also <span class="application">libpq</span> connection options, an error occurred.
     </p></li><li class="listitem"><p>
      Fix portability problems in <code class="filename">contrib/pageinspect</code>'s
      functions for GIN indexes (Peter Eisentraut, Tom Lane)
     </p></li><li class="listitem"><p>
      On Windows, ensure that environment variable changes are propagated
      to DLLs built with debug options (Christian Ullrich)
     </p></li><li class="listitem"><p>
      Sync our copy of the timezone library with IANA release tzcode2016j
      (Tom Lane)
     </p><p>
      This fixes various issues, most notably that timezone data
      installation failed if the target directory didn't support hard
      links.
     </p></li><li class="listitem"><p>
      Update time zone data files to <span class="application">tzdata</span> release 2016j
      for DST law changes in northern Cyprus (adding a new zone
      Asia/Famagusta), Russia (adding a new zone Europe/Saratov), Tonga,
      and Antarctica/Casey.
      Historical corrections for Italy, Kazakhstan, Malta, and Palestine.
      Switch to preferring numeric zone abbreviations for Tonga.
     </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-9-5-7.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-9-5-5.html">Next</a></td></tr><tr><td width="40%" align="left" valign="top">E.30. Release 9.5.7 </td><td width="20%" align="center"><a accesskey="h" href="index.html">Home</a></td><td width="40%" align="right" valign="top"> E.32. Release 9.5.5</td></tr></table></div></body></html>