~ubuntu-branches/ubuntu/trusty/postgresql-9.3/trusty-updates : contents of doc/src/sgml/html/release-9-3-3.html at revision 27

~ubuntu-branches/ubuntu/trusty/postgresql-9.3/trusty-updates : (revision 27)
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<HTML
><HEAD
><TITLE
>Release 9.3.3</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 9.3.13 Documentation"
HREF="index.html"><LINK
REL="UP"
TITLE="Release Notes"
HREF="release.html"><LINK
REL="PREVIOUS"
TITLE="Release 9.3.4"
HREF="release-9-3-4.html"><LINK
REL="NEXT"
TITLE="Release 9.3.2"
HREF="release-9-3-2.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="2016-05-09T21:13:26"></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 9.3.13 Documentation</A
></TH
></TR
><TR
><TD
WIDTH="10%"
ALIGN="left"
VALIGN="top"
><A
TITLE="Release 9.3.4"
HREF="release-9-3-4.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 9.3.2"
HREF="release-9-3-2.html"
ACCESSKEY="N"
>Next</A
></TD
></TR
></TABLE
><HR
ALIGN="LEFT"
WIDTH="100%"></DIV
><DIV
CLASS="SECT1"
><H1
CLASS="SECT1"
><A
NAME="RELEASE-9-3-3"
>E.11. Release 9.3.3</A
></H1
><DIV
CLASS="NOTE"
><BLOCKQUOTE
CLASS="NOTE"
><P
><B
>Release Date: </B
>2014-02-20</P
></BLOCKQUOTE
></DIV
><P
>   This release contains a variety of fixes from 9.3.2.
   For information about new features in the 9.3 major release, see
   <A
HREF="release-9-3.html"
>Section E.14</A
>.
  </P
><DIV
CLASS="SECT2"
><H2
CLASS="SECT2"
><A
NAME="AEN116377"
>E.11.1. Migration to Version 9.3.3</A
></H2
><P
>    A dump/restore is not required for those running 9.3.X.
   </P
><P
>    However, several of the issues corrected in this release could have
    resulted in corruption of foreign-key constraints; that is, there
    might now be referencing rows for which there is no matching row in
    the referenced table.  It may be worthwhile to recheck such
    constraints after installing this update.  The simplest way to do that
    is to drop and recreate each suspect constraint; however, that will
    require taking an exclusive lock on both tables, so it is unlikely to
    be acceptable in production databases.  Alternatively, you can do a
    manual join query between the two tables to look for unmatched rows.
   </P
><P
>    Note also the requirement for replication standby servers to be
    upgraded before their master server is upgraded.
   </P
><P
>    Also, if you are upgrading from a version earlier than 9.3.2,
    see <A
HREF="release-9-3-2.html"
>Section E.12</A
>.
   </P
></DIV
><DIV
CLASS="SECT2"
><H2
CLASS="SECT2"
><A
NAME="AEN116384"
>E.11.2. Changes</A
></H2
><P
></P
><UL
><LI
><P
>      Shore up <TT
CLASS="LITERAL"
>GRANT ... WITH ADMIN OPTION</TT
> restrictions
      (Noah Misch)
     </P
><P
>      Granting a role without <TT
CLASS="LITERAL"
>ADMIN OPTION</TT
> is supposed to
      prevent the grantee from adding or removing members from the granted
      role, but this restriction was easily bypassed by doing <TT
CLASS="LITERAL"
>SET
      ROLE</TT
> first.  The security impact is mostly that a role member can
      revoke the access of others, contrary to the wishes of his grantor.
      Unapproved role member additions are a lesser concern, since an
      uncooperative role member could provide most of his rights to others
      anyway by creating views or <TT
CLASS="LITERAL"
>SECURITY DEFINER</TT
> functions.
      (CVE-2014-0060)
     </P
></LI
><LI
><P
>      Prevent privilege escalation via manual calls to PL validator
      functions (Andres Freund)
     </P
><P
>      The primary role of PL validator functions is to be called implicitly
      during <TT
CLASS="COMMAND"
>CREATE FUNCTION</TT
>, but they are also normal SQL
      functions that a user can call explicitly.  Calling a validator on
      a function actually written in some other language was not checked
      for and could be exploited for privilege-escalation purposes.
      The fix involves adding a call to a privilege-checking function in
      each validator function.  Non-core procedural languages will also
      need to make this change to their own validator functions, if any.
      (CVE-2014-0061)
     </P
></LI
><LI
><P
>      Avoid multiple name lookups during table and index DDL
      (Robert Haas, Andres Freund)
     </P
><P
>      If the name lookups come to different conclusions due to concurrent
      activity, we might perform some parts of the DDL on a different table
      than other parts.  At least in the case of <TT
CLASS="COMMAND"
>CREATE INDEX</TT
>,
      this can be used to cause the permissions checks to be performed
      against a different table than the index creation, allowing for a
      privilege escalation attack.
      (CVE-2014-0062)
     </P
></LI
><LI
><P
>      Prevent buffer overrun with long datetime strings (Noah Misch)
     </P
><P
>      The <TT
CLASS="LITERAL"
>MAXDATELEN</TT
> constant was too small for the longest
      possible value of type <TT
CLASS="TYPE"
>interval</TT
>, allowing a buffer overrun
      in <CODE
CLASS="FUNCTION"
>interval_out()</CODE
>.  Although the datetime input
      functions were more careful about avoiding buffer overrun, the limit
      was short enough to cause them to reject some valid inputs, such as
      input containing a very long timezone name.  The <SPAN
CLASS="APPLICATION"
>ecpg</SPAN
>
      library contained these vulnerabilities along with some of its own.
      (CVE-2014-0063)
     </P
></LI
><LI
><P
>      Prevent buffer overrun due to integer overflow in size calculations
      (Noah Misch, Heikki Linnakangas)
     </P
><P
>      Several functions, mostly type input functions, calculated an
      allocation size without checking for overflow.  If overflow did
      occur, a too-small buffer would be allocated and then written past.
      (CVE-2014-0064)
     </P
></LI
><LI
><P
>      Prevent overruns of fixed-size buffers
      (Peter Eisentraut, Jozef Mlich)
     </P
><P
>      Use <CODE
CLASS="FUNCTION"
>strlcpy()</CODE
> and related functions to provide a clear
      guarantee that fixed-size buffers are not overrun.  Unlike the
      preceding items, it is unclear whether these cases really represent
      live issues, since in most cases there appear to be previous
      constraints on the size of the input string.  Nonetheless it seems
      prudent to silence all Coverity warnings of this type.
      (CVE-2014-0065)
     </P
></LI
><LI
><P
>      Avoid crashing if <CODE
CLASS="FUNCTION"
>crypt()</CODE
> returns NULL (Honza Horak,
      Bruce Momjian)
     </P
><P
>      There are relatively few scenarios in which <CODE
CLASS="FUNCTION"
>crypt()</CODE
>
      could return NULL, but <TT
CLASS="FILENAME"
>contrib/chkpass</TT
> would crash
      if it did.  One practical case in which this could be an issue is
      if <SPAN
CLASS="APPLICATION"
>libc</SPAN
> is configured to refuse to execute unapproved
      hashing algorithms (e.g., <SPAN
CLASS="QUOTE"
>"FIPS mode"</SPAN
>).
      (CVE-2014-0066)
     </P
></LI
><LI
><P
>      Document risks of <TT
CLASS="LITERAL"
>make check</TT
> in the regression testing
      instructions (Noah Misch, Tom Lane)
     </P
><P
>      Since the temporary server started by <TT
CLASS="LITERAL"
>make check</TT
>
      uses <SPAN
CLASS="QUOTE"
>"trust"</SPAN
> authentication, another user on the same machine
      could connect to it as database superuser, and then potentially
      exploit the privileges of the operating-system user who started the
      tests.  A future release will probably incorporate changes in the
      testing procedure to prevent this risk, but some public discussion is
      needed first.  So for the moment, just warn people against using
      <TT
CLASS="LITERAL"
>make check</TT
> when there are untrusted users on the
      same machine.
      (CVE-2014-0067)
     </P
></LI
><LI
><P
>      Rework tuple freezing protocol
      (&Aacute;lvaro Herrera, Andres Freund)
     </P
><P
>      The logic for tuple freezing was unable to handle some cases involving
      freezing of
      <A
HREF="routine-vacuuming.html#VACUUM-FOR-MULTIXACT-WRAPAROUND"
><I
CLASS="FIRSTTERM"
>multixact</I
>
      IDs</A
>, with the practical effect that shared row-level locks
      might be forgotten once old enough.
     </P
><P
>      Fixing this required changing the WAL record format for tuple
      freezing.  While this is no issue for standalone servers, when using
      replication it means that <SPAN
CLASS="emphasis"
><I
CLASS="EMPHASIS"
>standby servers must be upgraded
      to 9.3.3 or later before their masters are</I
></SPAN
>.  An older standby will
      be unable to interpret freeze records generated by a newer master, and
      will fail with a PANIC message.  (In such a case, upgrading the
      standby should be sufficient to let it resume execution.)
     </P
></LI
><LI
><P
>      Create separate GUC parameters to control multixact freezing
      (&Aacute;lvaro Herrera)
     </P
><P
>      9.3 requires multixact tuple labels to be frozen before
      they grow too old, in the same fashion as plain transaction ID labels
      have been frozen for some time.  Previously, the transaction ID
      freezing parameters were used for multixact IDs too; but since
      the consumption rates of transaction IDs and multixact IDs can be
      quite different, this did not work very well.  Introduce new settings
      <A
HREF="runtime-config-client.html#GUC-VACUUM-MULTIXACT-FREEZE-MIN-AGE"
>vacuum_multixact_freeze_min_age</A
>,
      <A
HREF="runtime-config-client.html#GUC-VACUUM-MULTIXACT-FREEZE-TABLE-AGE"
>vacuum_multixact_freeze_table_age</A
>, and
      <A
HREF="runtime-config-autovacuum.html#GUC-AUTOVACUUM-MULTIXACT-FREEZE-MAX-AGE"
>autovacuum_multixact_freeze_max_age</A
>
      to control when to freeze multixacts.
     </P
></LI
><LI
><P
>      Account for remote row locks propagated by local updates
      (&Aacute;lvaro Herrera)
     </P
><P
>      If a row was locked by transaction A, and transaction B updated it,
      the new version of the row created by B would be locked by A, yet
      visible only to B.  If transaction B then again updated the row, A's
      lock wouldn't get checked, thus possibly allowing B to complete when
      it shouldn't.  This case is new in 9.3 since prior versions did not
      have any types of row locking that would permit another transaction
      to update the row at all.
     </P
><P
>      This oversight could allow referential integrity checks to give false
      positives (for instance, allow deletes that should have been rejected).
      Applications using the new commands <TT
CLASS="LITERAL"
>SELECT FOR KEY SHARE</TT
>
      and <TT
CLASS="LITERAL"
>SELECT FOR NO KEY UPDATE</TT
> might also have suffered
      locking failures of this kind.
     </P
></LI
><LI
><P
>      Prevent <SPAN
CLASS="QUOTE"
>"forgetting"</SPAN
> valid row locks when one of several
      holders of a row lock aborts (&Aacute;lvaro Herrera)
     </P
><P
>      This was yet another mechanism by which a shared row lock could be
      lost, thus possibly allowing updates that should have been prevented
      by foreign-key constraints.
     </P
></LI
><LI
><P
>      Fix incorrect logic during update chain locking
      (&Aacute;lvaro Herrera)
     </P
><P
>      This mistake could result in spurious <SPAN
CLASS="QUOTE"
>"could not serialize access
      due to concurrent update"</SPAN
> errors in <TT
CLASS="LITERAL"
>REPEATABLE READ</TT
>
      and <TT
CLASS="LITERAL"
>SERIALIZABLE</TT
> transaction isolation modes.
     </P
></LI
><LI
><P
>      Handle wraparound correctly during extension or truncation
      of <TT
CLASS="FILENAME"
>pg_multixact/members</TT
>
      (Andres Freund, &Aacute;lvaro Herrera)
     </P
></LI
><LI
><P
>      Fix handling of 5-digit filenames in <TT
CLASS="FILENAME"
>pg_multixact/members</TT
>
      (&Aacute;lvaro Herrera)
     </P
><P
>      As of 9.3, these names can be more than 4 digits, but the directory
      cleanup code ignored such files.
     </P
></LI
><LI
><P
>      Improve performance of multixact cache code
      (&Aacute;lvaro Herrera)
     </P
></LI
><LI
><P
>      Optimize updating a row that's already locked by the same transaction
      (Andres Freund, &Aacute;lvaro Herrera)
     </P
><P
>      This fixes a performance regression from pre-9.3 versions when doing
      <TT
CLASS="LITERAL"
>SELECT FOR UPDATE</TT
> followed by <TT
CLASS="LITERAL"
>UPDATE/DELETE</TT
>.
     </P
></LI
><LI
><P
>      During archive recovery, prefer highest timeline number when WAL
      segments with the same ID are present in both the archive
      and <TT
CLASS="FILENAME"
>pg_xlog/</TT
> (Kyotaro Horiguchi)
     </P
><P
>      Previously, not-yet-archived segments could get ignored during
      recovery.  This reverts an undesirable behavioral change in 9.3.0
      back to the way things worked pre-9.3.
     </P
></LI
><LI
><P
>      Fix possible mis-replay of WAL records when some segments of a
      relation aren't full size (Greg Stark, Tom Lane)
     </P
><P
>      The WAL update could be applied to the wrong page, potentially many
      pages past where it should have been.  Aside from corrupting data,
      this error has been observed to result in significant <SPAN
CLASS="QUOTE"
>"bloat"</SPAN
>
      of standby servers compared to their masters, due to updates being
      applied far beyond where the end-of-file should have been.  This
      failure mode does not appear to be a significant risk during crash
      recovery, only when initially synchronizing a standby created from a
      base backup taken from a quickly-changing master.
     </P
></LI
><LI
><P
>      Fix bug in determining when recovery has reached consistency
      (Tomonari Katsumata, Heikki Linnakangas)
     </P
><P
>      In some cases WAL replay would mistakenly conclude that the database
      was already consistent at the start of replay, thus possibly allowing
      hot-standby queries before the database was really consistent.  Other
      symptoms such as <SPAN
CLASS="QUOTE"
>"PANIC: WAL contains references to invalid
      pages"</SPAN
> were also possible.
     </P
></LI
><LI
><P
>      Fix WAL logging of visibility map changes (Heikki Linnakangas)
     </P
></LI
><LI
><P
>      Fix improper locking of btree index pages while replaying
      a <TT
CLASS="LITERAL"
>VACUUM</TT
> operation in hot-standby mode (Andres Freund,
      Heikki Linnakangas, Tom Lane)
     </P
><P
>      This error could result in <SPAN
CLASS="QUOTE"
>"PANIC: WAL contains references to
      invalid pages"</SPAN
> failures.
     </P
></LI
><LI
><P
>      Ensure that insertions into non-leaf GIN index pages write a full-page
      WAL record when appropriate (Heikki Linnakangas)
     </P
><P
>      The previous coding risked index corruption in the event of a
      partial-page write during a system crash.
     </P
></LI
><LI
><P
>      When <TT
CLASS="LITERAL"
>pause_at_recovery_target</TT
>
      and <TT
CLASS="LITERAL"
>recovery_target_inclusive</TT
> are both set, ensure the
      target record is applied before pausing, not after (Heikki
      Linnakangas)
     </P
></LI
><LI
><P
>      Ensure walreceiver sends hot-standby feedback messages on time even
      when there is a continuous stream of data (Andres Freund, Amit
      Kapila)
     </P
></LI
><LI
><P
>      Prevent timeout interrupts from taking control away from mainline
      code unless <TT
CLASS="VARNAME"
>ImmediateInterruptOK</TT
> is set
      (Andres Freund, Tom Lane)
     </P
><P
>      This is a serious issue for any application making use of statement
      timeouts, as it could cause all manner of strange failures after a
      timeout occurred.  We have seen reports of <SPAN
CLASS="QUOTE"
>"stuck"</SPAN
> spinlocks,
      ERRORs being unexpectedly promoted to PANICs, unkillable backends,
      and other misbehaviors.
     </P
></LI
><LI
><P
>      Fix race conditions during server process exit (Robert Haas)
     </P
><P
>      Ensure that signal handlers don't attempt to use the
      process's <TT
CLASS="VARNAME"
>MyProc</TT
> pointer after it's no longer valid.
     </P
></LI
><LI
><P
>      Fix race conditions in walsender shutdown logic and walreceiver
      SIGHUP signal handler (Tom Lane)
     </P
></LI
><LI
><P
>      Fix unsafe references to <TT
CLASS="VARNAME"
>errno</TT
> within error reporting
      logic (Christian Kruse)
     </P
><P
>      This would typically lead to odd behaviors such as missing or
      inappropriate <TT
CLASS="LITERAL"
>HINT</TT
> fields.
     </P
></LI
><LI
><P
>      Fix possible crashes from using <CODE
CLASS="FUNCTION"
>ereport()</CODE
> too early
      during server startup (Tom Lane)
     </P
><P
>      The principal case we've seen in the field is a crash if the server
      is started in a directory it doesn't have permission to read.
     </P
></LI
><LI
><P
>      Clear retry flags properly in OpenSSL socket write
      function (Alexander Kukushkin)
     </P
><P
>      This omission could result in a server lockup after unexpected loss
      of an SSL-encrypted connection.
     </P
></LI
><LI
><P
>      Fix length checking for Unicode identifiers (<TT
CLASS="LITERAL"
>U&amp;"..."</TT
>
      syntax) containing escapes (Tom Lane)
     </P
><P
>      A spurious truncation warning would be printed for such identifiers
      if the escaped form of the identifier was too long, but the
      identifier actually didn't need truncation after de-escaping.
     </P
></LI
><LI
><P
>      Fix parsing of Unicode literals and identifiers just before the end
      of a command string or function body (Tom Lane)
     </P
></LI
><LI
><P
>      Allow keywords that are type names to be used in lists of roles
      (Stephen Frost)
     </P
><P
>      A previous patch allowed such keywords to be used without quoting
      in places such as role identifiers; but it missed cases where a
      list of role identifiers was permitted, such as <TT
CLASS="LITERAL"
>DROP ROLE</TT
>.
     </P
></LI
><LI
><P
>      Fix parser crash for <TT
CLASS="LITERAL"
>EXISTS(SELECT * FROM
      zero_column_table)</TT
> (Tom Lane)
     </P
></LI
><LI
><P
>      Fix possible crash due to invalid plan for nested sub-selects, such
      as <TT
CLASS="LITERAL"
>WHERE (... x IN (SELECT ...) ...) IN (SELECT ...)</TT
>
      (Tom Lane)
     </P
></LI
><LI
><P
>      Fix mishandling of <TT
CLASS="LITERAL"
>WHERE</TT
> conditions pulled up from
      a <TT
CLASS="LITERAL"
>LATERAL</TT
> subquery (Tom Lane)
     </P
><P
>      The typical symptom of this bug was a <SPAN
CLASS="QUOTE"
>"JOIN qualification
      cannot refer to other relations"</SPAN
> error, though subtle logic
      errors in created plans seem possible as well.
     </P
></LI
><LI
><P
>      Disallow <TT
CLASS="LITERAL"
>LATERAL</TT
> references to the target table of
      an <TT
CLASS="LITERAL"
>UPDATE/DELETE</TT
> (Tom Lane)
     </P
><P
>      While this might be allowed in some future release, it was
      unintentional in 9.3, and didn't work quite right anyway.
     </P
></LI
><LI
><P
>      Fix <TT
CLASS="LITERAL"
>UPDATE/DELETE</TT
> of an inherited target table
      that has <TT
CLASS="LITERAL"
>UNION ALL</TT
> subqueries (Tom Lane)
     </P
><P
>      Without this fix, <TT
CLASS="LITERAL"
>UNION ALL</TT
> subqueries aren't correctly
      inserted into the update plans for inheritance child tables after the
      first one, typically resulting in no update happening for those child
      table(s).
     </P
></LI
><LI
><P
>      Fix <TT
CLASS="COMMAND"
>ANALYZE</TT
> to not fail on a column that's a domain over
      a range type (Tom Lane)
     </P
></LI
><LI
><P
>      Ensure that <TT
CLASS="COMMAND"
>ANALYZE</TT
> creates statistics for a table column
      even when all the values in it are <SPAN
CLASS="QUOTE"
>"too wide"</SPAN
> (Tom Lane)
     </P
><P
>      <TT
CLASS="COMMAND"
>ANALYZE</TT
> intentionally omits very wide values from its
      histogram and most-common-values calculations, but it neglected to do
      something sane in the case that all the sampled entries are too wide.
     </P
></LI
><LI
><P
>      In <TT
CLASS="LITERAL"
>ALTER TABLE ... SET TABLESPACE</TT
>, allow the database's
      default tablespace to be used without a permissions check
      (Stephen Frost)
     </P
><P
>      <TT
CLASS="LITERAL"
>CREATE TABLE</TT
> has always allowed such usage,
      but <TT
CLASS="LITERAL"
>ALTER TABLE</TT
> didn't get the memo.
     </P
></LI
><LI
><P
>      Fix support for extensions containing event triggers (Tom Lane)
     </P
></LI
><LI
><P
>      Fix <SPAN
CLASS="QUOTE"
>"cannot accept a set"</SPAN
> error when some arms of
      a <TT
CLASS="LITERAL"
>CASE</TT
> return a set and others don't (Tom Lane)
     </P
></LI
><LI
><P
>      Fix memory leakage in JSON functions (Craig Ringer)
     </P
></LI
><LI
><P
>      Properly distinguish numbers from non-numbers when generating JSON
      output (Andrew Dunstan)
     </P
></LI
><LI
><P
>      Fix checks for all-zero client addresses in pgstat functions (Kevin
      Grittner)
     </P
></LI
><LI
><P
>      Fix possible misclassification of multibyte characters by the text
      search parser (Tom Lane)
     </P
><P
>      Non-ASCII characters could be misclassified when using C locale with
      a multibyte encoding.  On Cygwin, non-C locales could fail as well.
     </P
></LI
><LI
><P
>      Fix possible misbehavior in <CODE
CLASS="FUNCTION"
>plainto_tsquery()</CODE
>
      (Heikki Linnakangas)
     </P
><P
>      Use <CODE
CLASS="FUNCTION"
>memmove()</CODE
> not <CODE
CLASS="FUNCTION"
>memcpy()</CODE
> for copying
      overlapping memory regions.  There have been no field reports of
      this actually causing trouble, but it's certainly risky.
     </P
></LI
><LI
><P
>      Fix placement of permissions checks in <CODE
CLASS="FUNCTION"
>pg_start_backup()</CODE
>
      and <CODE
CLASS="FUNCTION"
>pg_stop_backup()</CODE
> (Andres Freund, Magnus Hagander)
     </P
><P
>      The previous coding might attempt to do catalog access when it
      shouldn't.
     </P
></LI
><LI
><P
>      Accept <TT
CLASS="LITERAL"
>SHIFT_JIS</TT
> as an encoding name for locale checking
      purposes (Tatsuo Ishii)
     </P
></LI
><LI
><P
>      Fix <TT
CLASS="LITERAL"
>*</TT
>-qualification of named parameters in SQL-language
      functions (Tom Lane)
     </P
><P
>      Given a composite-type parameter
      named <TT
CLASS="LITERAL"
>foo</TT
>, <TT
CLASS="LITERAL"
>$1.*</TT
> worked fine,
      but <TT
CLASS="LITERAL"
>foo.*</TT
> not so much.
     </P
></LI
><LI
><P
>      Fix misbehavior of <CODE
CLASS="FUNCTION"
>PQhost()</CODE
> on Windows (Fujii Masao)
     </P
><P
>      It should return <TT
CLASS="LITERAL"
>localhost</TT
> if no host has been specified.
     </P
></LI
><LI
><P
>      Improve error handling in <SPAN
CLASS="APPLICATION"
>libpq</SPAN
> and <SPAN
CLASS="APPLICATION"
>psql</SPAN
>
      for failures during <TT
CLASS="LITERAL"
>COPY TO STDOUT/FROM STDIN</TT
> (Tom Lane)
     </P
><P
>      In particular this fixes an infinite loop that could occur in 9.2 and
      up if the server connection was lost during <TT
CLASS="LITERAL"
>COPY FROM
      STDIN</TT
>.  Variants of that scenario might be possible in older
      versions, or with other client applications.
     </P
></LI
><LI
><P
>      Fix incorrect translation handling in
      some <SPAN
CLASS="APPLICATION"
>psql</SPAN
> <TT
CLASS="LITERAL"
>\d</TT
> commands
      (Peter Eisentraut, Tom Lane)
     </P
></LI
><LI
><P
>      Ensure <SPAN
CLASS="APPLICATION"
>pg_basebackup</SPAN
>'s background process is killed
      when exiting its foreground process (Magnus Hagander)
     </P
></LI
><LI
><P
>      Fix possible incorrect printing of filenames
      in <SPAN
CLASS="APPLICATION"
>pg_basebackup</SPAN
>'s verbose mode (Magnus Hagander)
     </P
></LI
><LI
><P
>      Avoid including tablespaces inside PGDATA twice in base backups
      (Dimitri Fontaine, Magnus Hagander)
     </P
></LI
><LI
><P
>      Fix misaligned descriptors in <SPAN
CLASS="APPLICATION"
>ecpg</SPAN
> (MauMau)
     </P
></LI
><LI
><P
>      In <SPAN
CLASS="APPLICATION"
>ecpg</SPAN
>, handle lack of a hostname in the connection
      parameters properly (Michael Meskes)
     </P
></LI
><LI
><P
>      Fix performance regression in <TT
CLASS="FILENAME"
>contrib/dblink</TT
> connection
      startup (Joe Conway)
     </P
><P
>      Avoid an unnecessary round trip when client and server encodings match.
     </P
></LI
><LI
><P
>      In <TT
CLASS="FILENAME"
>contrib/isn</TT
>, fix incorrect calculation of the check
      digit for ISMN values (Fabien Coelho)
     </P
></LI
><LI
><P
>      Fix <TT
CLASS="FILENAME"
>contrib/pgbench</TT
>'s progress logging to avoid overflow
      when the scale factor is large (Tatsuo Ishii)
     </P
></LI
><LI
><P
>      Fix <TT
CLASS="FILENAME"
>contrib/pg_stat_statement</TT
>'s handling
      of <TT
CLASS="LITERAL"
>CURRENT_DATE</TT
> and related constructs (Kyotaro
      Horiguchi)
     </P
></LI
><LI
><P
>      Improve lost-connection error handling
      in <TT
CLASS="FILENAME"
>contrib/postgres_fdw</TT
> (Tom Lane)
     </P
></LI
><LI
><P
>      Ensure client-code-only installation procedure works as documented
      (Peter Eisentraut)
     </P
></LI
><LI
><P
>      In Mingw and Cygwin builds, install the <SPAN
CLASS="APPLICATION"
>libpq</SPAN
> DLL
      in the <TT
CLASS="FILENAME"
>bin</TT
> directory (Andrew Dunstan)
     </P
><P
>      This duplicates what the MSVC build has long done.  It should fix
      problems with programs like <SPAN
CLASS="APPLICATION"
>psql</SPAN
> failing to start
      because they can't find the DLL.
     </P
></LI
><LI
><P
>      Avoid using the deprecated <TT
CLASS="LITERAL"
>dllwrap</TT
> tool in Cygwin builds
      (Marco Atzeri)
     </P
></LI
><LI
><P
>      Enable building with Visual Studio 2013 (Brar Piening)
     </P
></LI
><LI
><P
>      Don't generate plain-text <TT
CLASS="FILENAME"
>HISTORY</TT
>
      and <TT
CLASS="FILENAME"
>src/test/regress/README</TT
> files anymore (Tom Lane)
     </P
><P
>      These text files duplicated the main HTML and PDF documentation
      formats.  The trouble involved in maintaining them greatly outweighs
      the likely audience for plain-text format.  Distribution tarballs
      will still contain files by these names, but they'll just be stubs
      directing the reader to consult the main documentation.
      The plain-text <TT
CLASS="FILENAME"
>INSTALL</TT
> file will still be maintained, as
      there is arguably a use-case for that.
     </P
></LI
><LI
><P
>      Update time zone data files to <SPAN
CLASS="APPLICATION"
>tzdata</SPAN
> release 2013i
      for DST law changes in Jordan and historical changes in Cuba.
     </P
><P
>      In addition, the zones <TT
CLASS="LITERAL"
>Asia/Riyadh87</TT
>,
      <TT
CLASS="LITERAL"
>Asia/Riyadh88</TT
>, and <TT
CLASS="LITERAL"
>Asia/Riyadh89</TT
> have been
      removed, as they are no longer maintained by IANA, and never
      represented actual civil timekeeping practice.
     </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-9-3-4.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-9-3-2.html"
ACCESSKEY="N"
>Next</A
></TD
></TR
><TR
><TD
WIDTH="33%"
ALIGN="left"
VALIGN="top"
>Release 9.3.4</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 9.3.2</TD
></TR
></TABLE
></DIV
></BODY
></HTML
>