1#@!#!123s

: / opt / alt / postgresql11 / usr / share / doc / alt-postgresql11-9.2.24 / html /
Filename : release-9-1-20.html
back
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<HTML
><HEAD
><TITLE
>Release 9.1.20</TITLE
><META
NAME="GENERATOR"
CONTENT="Modular DocBook HTML Stylesheet Version 1.79"><LINK
REV="MADE"
HREF="mailto:pgsql-docs@postgresql.org"><LINK
REL="HOME"
TITLE="PostgreSQL 9.2.24 Documentation"
HREF="index.html"><LINK
REL="UP"
TITLE="Release Notes"
HREF="release.html"><LINK
REL="PREVIOUS"
TITLE="Release 9.1.21"
HREF="release-9-1-21.html"><LINK
REL="NEXT"
TITLE="Release 9.1.19"
HREF="release-9-1-19.html"><LINK
REL="STYLESHEET"
TYPE="text/css"
HREF="stylesheet.css"><META
HTTP-EQUIV="Content-Type"
CONTENT="text/html; charset=ISO-8859-1"><META
NAME="creation"
CONTENT="2017-11-06T22:43:11"></HEAD
><BODY
CLASS="SECT1"
><DIV
CLASS="NAVHEADER"
><TABLE
SUMMARY="Header navigation table"
WIDTH="100%"
BORDER="0"
CELLPADDING="0"
CELLSPACING="0"
><TR
><TH
COLSPAN="5"
ALIGN="center"
VALIGN="bottom"
><A
HREF="index.html"
>PostgreSQL 9.2.24 Documentation</A
></TH
></TR
><TR
><TD
WIDTH="10%"
ALIGN="left"
VALIGN="top"
><A
TITLE="Release 9.1.21"
HREF="release-9-1-21.html"
ACCESSKEY="P"
>Prev</A
></TD
><TD
WIDTH="10%"
ALIGN="left"
VALIGN="top"
><A
HREF="release.html"
ACCESSKEY="U"
>Up</A
></TD
><TD
WIDTH="60%"
ALIGN="center"
VALIGN="bottom"
>Appendix E. Release Notes</TD
><TD
WIDTH="20%"
ALIGN="right"
VALIGN="top"
><A
TITLE="Release 9.1.19"
HREF="release-9-1-19.html"
ACCESSKEY="N"
>Next</A
></TD
></TR
></TABLE
><HR
ALIGN="LEFT"
WIDTH="100%"></DIV
><DIV
CLASS="SECT1"
><H1
CLASS="SECT1"
><A
NAME="RELEASE-9-1-20"
>E.30. Release 9.1.20</A
></H1
><DIV
CLASS="FORMALPARA"
><P
><B
>Release date: </B
>2016-02-11</P
></DIV
><P
>   This release contains a variety of fixes from 9.1.19.
   For information about new features in the 9.1 major release, see
   <A
HREF="release-9-1.html"
>Section E.50</A
>.
  </P
><DIV
CLASS="SECT2"
><H2
CLASS="SECT2"
><A
NAME="AEN116031"
>E.30.1. Migration to Version 9.1.20</A
></H2
><P
>    A dump/restore is not required for those running 9.1.X.
   </P
><P
>    However, if you are upgrading from a version earlier than 9.1.16,
    see <A
HREF="release-9-1-16.html"
>Section E.34</A
>.
   </P
></DIV
><DIV
CLASS="SECT2"
><H2
CLASS="SECT2"
><A
NAME="AEN116036"
>E.30.2. Changes</A
></H2
><P
></P
><UL
><LI
><P
>      Fix infinite loops and buffer-overrun problems in regular expressions
      (Tom Lane)
     </P
><P
>      Very large character ranges in bracket expressions could cause
      infinite loops in some cases, and memory overwrites in other cases.
      (CVE-2016-0773)
     </P
></LI
><LI
><P
>      Perform an immediate shutdown if the <TT
CLASS="FILENAME"
>postmaster.pid</TT
> file
      is removed (Tom Lane)
     </P
><P
>      The postmaster now checks every minute or so
      that <TT
CLASS="FILENAME"
>postmaster.pid</TT
> is still there and still contains its
      own PID.  If not, it performs an immediate shutdown, as though it had
      received <SPAN
CLASS="SYSTEMITEM"
>SIGQUIT</SPAN
>.  The main motivation for this change
      is to ensure that failed buildfarm runs will get cleaned up without
      manual intervention; but it also serves to limit the bad effects if a
      DBA forcibly removes <TT
CLASS="FILENAME"
>postmaster.pid</TT
> and then starts a new
      postmaster.
     </P
></LI
><LI
><P
>      In <TT
CLASS="LITERAL"
>SERIALIZABLE</TT
> transaction isolation mode, serialization
      anomalies could be missed due to race conditions during insertions
      (Kevin Grittner, Thomas Munro)
     </P
></LI
><LI
><P
>      Fix failure to emit appropriate WAL records when doing <TT
CLASS="LITERAL"
>ALTER
      TABLE ... SET TABLESPACE</TT
> for unlogged relations (Michael Paquier,
      Andres Freund)
     </P
><P
>      Even though the relation's data is unlogged, the move must be logged or
      the relation will be inaccessible after a standby is promoted to master.
     </P
></LI
><LI
><P
>      Fix possible misinitialization of unlogged relations at the end of
      crash recovery (Andres Freund, Michael Paquier)
     </P
></LI
><LI
><P
>      Fix <TT
CLASS="COMMAND"
>ALTER COLUMN TYPE</TT
> to reconstruct inherited check
      constraints properly (Tom Lane)
     </P
></LI
><LI
><P
>      Fix <TT
CLASS="COMMAND"
>REASSIGN OWNED</TT
> to change ownership of composite types
      properly (&Aacute;lvaro Herrera)
     </P
></LI
><LI
><P
>      Fix <TT
CLASS="COMMAND"
>REASSIGN OWNED</TT
> and <TT
CLASS="COMMAND"
>ALTER OWNER</TT
> to correctly
      update granted-permissions lists when changing owners of data types,
      foreign data wrappers, or foreign servers (Bruce Momjian,
      &Aacute;lvaro Herrera)
     </P
></LI
><LI
><P
>      Fix <TT
CLASS="COMMAND"
>REASSIGN OWNED</TT
> to ignore foreign user mappings,
      rather than fail (&Aacute;lvaro Herrera)
     </P
></LI
><LI
><P
>      Add more defenses against bad planner cost estimates for GIN index
      scans when the index's internal statistics are very out-of-date
      (Tom Lane)
     </P
></LI
><LI
><P
>      Make planner cope with hypothetical GIN indexes suggested by an index
      advisor plug-in (Julien Rouhaud)
     </P
></LI
><LI
><P
>      Fix dumping of whole-row Vars in <TT
CLASS="LITERAL"
>ROW()</TT
>
      and <TT
CLASS="LITERAL"
>VALUES()</TT
> lists (Tom Lane)
     </P
></LI
><LI
><P
>      Fix possible internal overflow in <TT
CLASS="TYPE"
>numeric</TT
> division
      (Dean Rasheed)
     </P
></LI
><LI
><P
>      Fix enforcement of restrictions inside parentheses within regular
      expression lookahead constraints (Tom Lane)
     </P
><P
>      Lookahead constraints aren't allowed to contain backrefs, and
      parentheses within them are always considered non-capturing, according
      to the manual.  However, the code failed to handle these cases properly
      inside a parenthesized subexpression, and would give unexpected
      results.
     </P
></LI
><LI
><P
>      Conversion of regular expressions to indexscan bounds could produce
      incorrect bounds from regexps containing lookahead constraints
      (Tom Lane)
     </P
></LI
><LI
><P
>      Fix regular-expression compiler to handle loops of constraint arcs
      (Tom Lane)
     </P
><P
>      The code added for CVE-2007-4772 was both incomplete, in that it didn't
      handle loops involving more than one state, and incorrect, in that it
      could cause assertion failures (though there seem to be no bad
      consequences of that in a non-assert build).  Multi-state loops would
      cause the compiler to run until the query was canceled or it reached
      the too-many-states error condition.
     </P
></LI
><LI
><P
>      Improve memory-usage accounting in regular-expression compiler
      (Tom Lane)
     </P
><P
>      This causes the code to emit <SPAN
CLASS="QUOTE"
>"regular expression is too
      complex"</SPAN
> errors in some cases that previously used unreasonable
      amounts of time and memory.
     </P
></LI
><LI
><P
>      Improve performance of regular-expression compiler (Tom Lane)
     </P
></LI
><LI
><P
>      Make <TT
CLASS="LITERAL"
>%h</TT
> and <TT
CLASS="LITERAL"
>%r</TT
> escapes
      in <TT
CLASS="VARNAME"
>log_line_prefix</TT
> work for messages emitted due
      to <TT
CLASS="VARNAME"
>log_connections</TT
> (Tom Lane)
     </P
><P
>      Previously, <TT
CLASS="LITERAL"
>%h</TT
>/<TT
CLASS="LITERAL"
>%r</TT
> started to work just after a
      new session had emitted the <SPAN
CLASS="QUOTE"
>"connection received"</SPAN
> log message;
      now they work for that message too.
     </P
></LI
><LI
><P
>      On Windows, ensure the shared-memory mapping handle gets closed in
      child processes that don't need it (Tom Lane, Amit Kapila)
     </P
><P
>      This oversight resulted in failure to recover from crashes
      whenever <TT
CLASS="VARNAME"
>logging_collector</TT
> is turned on.
     </P
></LI
><LI
><P
>      Fix possible failure to detect socket EOF in non-blocking mode on
      Windows (Tom Lane)
     </P
><P
>      It's not entirely clear whether this problem can happen in pre-9.5
      branches, but if it did, the symptom would be that a walsender process
      would wait indefinitely rather than noticing a loss of connection.
     </P
></LI
><LI
><P
>      Avoid leaking a token handle during SSPI authentication
      (Christian Ullrich)
     </P
></LI
><LI
><P
>      In <SPAN
CLASS="APPLICATION"
>psql</SPAN
>, ensure that <SPAN
CLASS="APPLICATION"
>libreadline</SPAN
>'s idea
      of the screen size is updated when the terminal window size changes
      (Merlin Moncure)
     </P
><P
>      Previously, <SPAN
CLASS="APPLICATION"
>libreadline</SPAN
> did not notice if the window
      was resized during query output, leading to strange behavior during
      later input of multiline queries.
     </P
></LI
><LI
><P
>      Fix <SPAN
CLASS="APPLICATION"
>psql</SPAN
>'s <TT
CLASS="LITERAL"
>\det</TT
> command to interpret its
      pattern argument the same way as other <TT
CLASS="LITERAL"
>\d</TT
> commands with
      potentially schema-qualified patterns do (Reece Hart)
     </P
></LI
><LI
><P
>      Avoid possible crash in <SPAN
CLASS="APPLICATION"
>psql</SPAN
>'s <TT
CLASS="LITERAL"
>\c</TT
> command
      when previous connection was via Unix socket and command specifies a
      new hostname and same username (Tom Lane)
     </P
></LI
><LI
><P
>      In <TT
CLASS="LITERAL"
>pg_ctl start -w</TT
>, test child process status directly
      rather than relying on heuristics (Tom Lane, Michael Paquier)
     </P
><P
>      Previously, <SPAN
CLASS="APPLICATION"
>pg_ctl</SPAN
> relied on an assumption that the new
      postmaster would always create <TT
CLASS="FILENAME"
>postmaster.pid</TT
> within five
      seconds.  But that can fail on heavily-loaded systems,
      causing <SPAN
CLASS="APPLICATION"
>pg_ctl</SPAN
> to report incorrectly that the
      postmaster failed to start.
     </P
><P
>      Except on Windows, this change also means that a <TT
CLASS="LITERAL"
>pg_ctl start
      -w</TT
> done immediately after another such command will now reliably
      fail, whereas previously it would report success if done within two
      seconds of the first command.
     </P
></LI
><LI
><P
>      In <TT
CLASS="LITERAL"
>pg_ctl start -w</TT
>, don't attempt to use a wildcard listen
      address to connect to the postmaster (Kondo Yuta)
     </P
><P
>      On Windows, <SPAN
CLASS="APPLICATION"
>pg_ctl</SPAN
> would fail to detect postmaster
      startup if <TT
CLASS="VARNAME"
>listen_addresses</TT
> is set to <TT
CLASS="LITERAL"
>0.0.0.0</TT
>
      or <TT
CLASS="LITERAL"
>::</TT
>, because it would try to use that value verbatim as
      the address to connect to, which doesn't work.  Instead assume
      that <TT
CLASS="LITERAL"
>127.0.0.1</TT
> or <TT
CLASS="LITERAL"
>::1</TT
>, respectively, is the
      right thing to use.
     </P
></LI
><LI
><P
>      In <SPAN
CLASS="APPLICATION"
>pg_ctl</SPAN
> on Windows, check service status to decide
      where to send output, rather than checking if standard output is a
      terminal (Michael Paquier)
     </P
></LI
><LI
><P
>      In <SPAN
CLASS="APPLICATION"
>pg_dump</SPAN
> and <SPAN
CLASS="APPLICATION"
>pg_basebackup</SPAN
>, adopt
      the GNU convention for handling tar-archive members exceeding 8GB
      (Tom Lane)
     </P
><P
>      The POSIX standard for <TT
CLASS="LITERAL"
>tar</TT
> file format does not allow
      archive member files to exceed 8GB, but most modern implementations
      of <SPAN
CLASS="APPLICATION"
>tar</SPAN
> support an extension that fixes that.  Adopt
      this extension so that <SPAN
CLASS="APPLICATION"
>pg_dump</SPAN
> with <TT
CLASS="OPTION"
>-Ft</TT
> no
      longer fails on tables with more than 8GB of data, and so
      that <SPAN
CLASS="APPLICATION"
>pg_basebackup</SPAN
> can handle files larger than 8GB.
      In addition, fix some portability issues that could cause failures for
      members between 4GB and 8GB on some platforms.  Potentially these
      problems could cause unrecoverable data loss due to unreadable backup
      files.
     </P
></LI
><LI
><P
>      Fix assorted corner-case bugs in <SPAN
CLASS="APPLICATION"
>pg_dump</SPAN
>'s processing
      of extension member objects (Tom Lane)
     </P
></LI
><LI
><P
>      Make <SPAN
CLASS="APPLICATION"
>pg_dump</SPAN
> mark a view's triggers as needing to be
      processed after its rule, to prevent possible failure during
      parallel <SPAN
CLASS="APPLICATION"
>pg_restore</SPAN
> (Tom Lane)
     </P
></LI
><LI
><P
>      Ensure that relation option values are properly quoted
      in <SPAN
CLASS="APPLICATION"
>pg_dump</SPAN
> (Kouhei Sutou, Tom Lane)
     </P
><P
>      A reloption value that isn't a simple identifier or number could lead
      to dump/reload failures due to syntax errors in CREATE statements
      issued by <SPAN
CLASS="APPLICATION"
>pg_dump</SPAN
>.  This is not an issue with any
      reloption currently supported by core <SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
>, but
      extensions could allow reloptions that cause the problem.
     </P
></LI
><LI
><P
>      Fix <SPAN
CLASS="APPLICATION"
>pg_upgrade</SPAN
>'s file-copying code to handle errors
      properly on Windows (Bruce Momjian)
     </P
></LI
><LI
><P
>      Install guards in <SPAN
CLASS="APPLICATION"
>pgbench</SPAN
> against corner-case overflow
      conditions during evaluation of script-specified division or modulo
      operators (Fabien Coelho, Michael Paquier)
     </P
></LI
><LI
><P
>      Prevent certain <SPAN
CLASS="APPLICATION"
>PL/Java</SPAN
> parameters from being set by
      non-superusers (Noah Misch)
     </P
><P
>      This change mitigates a <SPAN
CLASS="APPLICATION"
>PL/Java</SPAN
> security bug
      (CVE-2016-0766), which was fixed in <SPAN
CLASS="APPLICATION"
>PL/Java</SPAN
> by marking
      these parameters as superuser-only.  To fix the security hazard for
      sites that update <SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
> more frequently
      than <SPAN
CLASS="APPLICATION"
>PL/Java</SPAN
>, make the core code aware of them also.
     </P
></LI
><LI
><P
>      Improve <SPAN
CLASS="APPLICATION"
>libpq</SPAN
>'s handling of out-of-memory situations
      (Michael Paquier, Amit Kapila, Heikki Linnakangas)
     </P
></LI
><LI
><P
>      Fix order of arguments
      in <SPAN
CLASS="APPLICATION"
>ecpg</SPAN
>-generated <TT
CLASS="LITERAL"
>typedef</TT
> statements
      (Michael Meskes)
     </P
></LI
><LI
><P
>      Use <TT
CLASS="LITERAL"
>%g</TT
> not <TT
CLASS="LITERAL"
>%f</TT
> format
      in <SPAN
CLASS="APPLICATION"
>ecpg</SPAN
>'s <CODE
CLASS="FUNCTION"
>PGTYPESnumeric_from_double()</CODE
>
      (Tom Lane)
     </P
></LI
><LI
><P
>      Fix <SPAN
CLASS="APPLICATION"
>ecpg</SPAN
>-supplied header files to not contain comments
      continued from a preprocessor directive line onto the next line
      (Michael Meskes)
     </P
><P
>      Such a comment is rejected by <SPAN
CLASS="APPLICATION"
>ecpg</SPAN
>.  It's not yet clear
      whether <SPAN
CLASS="APPLICATION"
>ecpg</SPAN
> itself should be changed.
     </P
></LI
><LI
><P
>      Ensure that <TT
CLASS="FILENAME"
>contrib/pgcrypto</TT
>'s <CODE
CLASS="FUNCTION"
>crypt()</CODE
>
      function can be interrupted by query cancel (Andreas Karlsson)
     </P
></LI
><LI
><P
>      Accept <SPAN
CLASS="APPLICATION"
>flex</SPAN
> versions later than 2.5.x
      (Tom Lane, Michael Paquier)
     </P
><P
>      Now that flex 2.6.0 has been released, the version checks in our build
      scripts needed to be adjusted.
     </P
></LI
><LI
><P
>      Install our <TT
CLASS="FILENAME"
>missing</TT
> script where PGXS builds can find it
      (Jim Nasby)
     </P
><P
>      This allows sane behavior in a PGXS build done on a machine where build
      tools such as <SPAN
CLASS="APPLICATION"
>bison</SPAN
> are missing.
     </P
></LI
><LI
><P
>      Ensure that <TT
CLASS="FILENAME"
>dynloader.h</TT
> is included in the installed
      header files in MSVC builds (Bruce Momjian, Michael Paquier)
     </P
></LI
><LI
><P
>      Add variant regression test expected-output file to match behavior of
      current <SPAN
CLASS="APPLICATION"
>libxml2</SPAN
> (Tom Lane)
     </P
><P
>      The fix for <SPAN
CLASS="APPLICATION"
>libxml2</SPAN
>'s CVE-2015-7499 causes it not to
      output error context reports in some cases where it used to do so.
      This seems to be a bug, but we'll probably have to live with it for
      some time, so work around it.
     </P
></LI
><LI
><P
>      Update time zone data files to <SPAN
CLASS="APPLICATION"
>tzdata</SPAN
> release 2016a for
      DST law changes in Cayman Islands, Metlakatla, and Trans-Baikal
      Territory (Zabaykalsky Krai), plus historical corrections for Pakistan.
     </P
></LI
></UL
></DIV
></DIV
><DIV
CLASS="NAVFOOTER"
><HR
ALIGN="LEFT"
WIDTH="100%"><TABLE
SUMMARY="Footer navigation table"
WIDTH="100%"
BORDER="0"
CELLPADDING="0"
CELLSPACING="0"
><TR
><TD
WIDTH="33%"
ALIGN="left"
VALIGN="top"
><A
HREF="release-9-1-21.html"
ACCESSKEY="P"
>Prev</A
></TD
><TD
WIDTH="34%"
ALIGN="center"
VALIGN="top"
><A
HREF="index.html"
ACCESSKEY="H"
>Home</A
></TD
><TD
WIDTH="33%"
ALIGN="right"
VALIGN="top"
><A
HREF="release-9-1-19.html"
ACCESSKEY="N"
>Next</A
></TD
></TR
><TR
><TD
WIDTH="33%"
ALIGN="left"
VALIGN="top"
>Release 9.1.21</TD
><TD
WIDTH="34%"
ALIGN="center"
VALIGN="top"
><A
HREF="release.html"
ACCESSKEY="U"
>Up</A
></TD
><TD
WIDTH="33%"
ALIGN="right"
VALIGN="top"
>Release 9.1.19</TD
></TR
></TABLE
></DIV
></BODY
></HTML
>