Skip to content

Wöchentlicher PostgreSQL Newsletter - 08. März 2009

Der Originalartikel befindet sich unter:

== Wöchentlicher PostgreSQL Newsletter - 08. März 2009 ==

Du kannst der pgsql-announce Mailingliste nun auf Twitter unter
@PGSQL_Announce folgen.

Kevin Kempter wird über Backup und Wiederherstellen auf dem ersten
Denver PUG Meeting am 8. April 2009 sprechen.

Neue Umfrage: Hast du jemals die PostgreSQL Systemkataloge genutzt?

Die Registrierung für die PostgreSQL Conference (JDCon) East 2009
ist jetzt offen.

== PostgreSQL Produkt Neuigkeiten ==

PostgreSQL Maestro 9.2, ein GUI zum Administrieren und Entwickeln unter
Windows, ist erschienen.

Npgsql2 2.0.4, ein .NET Datenprovider für PostgreSQL, ist erschienen.

Turnkey Linux PostgreSQL Appliance 8.3.5 erschienen.

Turnkey Linux LAPP Appliance erschienen.

Golconde 0.4, ein Queiebasiertes Replikationssystem in Python, ist

== PostgreSQL 8.4 Feature der Woche ==

Auto Explain. Dieses Contrib-Modul, wen installiert, erlaubt das An- und
Ausschalten (zur Laufzeit) der automatischen Explains für langsame
Anfragen in Postgres. Die Ausgabe wird geloggt.

== PostgreSQL Tipp der Woche ==

Zeilen in pg_hba.conf werden der Reihe nach abgearbeitet, nur die erste
passende Zeile wird in Betracht gezogen.

== PostgreSQL Jobs im März ==

== PostgreSQL Lokal ==

Die Deutsche PostgreSQL User Gruppe wird einen Stand, 2 Vorträge und
einen Workshop auf den Chemnitzer Linuxtagen am 14. und 15. März 2009
haben. Mehr Informationen auf Deutsch hier:

PostgreSQLFr wird einen Stand auf der Solutions Linux 2009 in Paris
haben. Anmelden unter sas AT postgresql DOT fr oder im
französischen Wiki.

PostgreSQL Conference U.S. wird einen PgDay auf dem Linuxfest Nordwest
am 25. und 26. April veranstalten. Der Call for Papers ist unter:

Michael Renner wird einen Workshop über PostgreSQL Replikation auf
der Netways OSDC 2009 in Nürnbern, Deutschland geben.

PGCon 2009 wird vom 21. bis 22. Mai 2009 in Ottawa an der Universität
von Ottawa stattfinden. Dem gehen zwei Tage mit Tutorials vom 19. bis
20. Mai 2009 voraus.

Termin vormerken: pgDay San Jose, am Sonntag dem 19. Juli unmittelbar
vor der OSCON. CfP und mehr Informationen bald.

PGCon Brasilien wird am 23.-24. Oktober 2009 auf dem Unicamp in
Campinas, Sao Paulo, stattfinden.

== PostgreSQL in den News ==

Planet PostgreSQL:

Dieser wöchentliche PostgreSQL Newsletter wurde erstellt von David
Fetter und Josh Berkus.

Sende Neuigkeiten und Ankündigungen bis Sonntag, 15 Uhr Pazifischer
Zeit. Bitte sende englische Beiträge an, deutsche an, italienische an

== Angewandte Patches ==

Teodor Sigaev committed:

- Fix usage of char2wchar/wchar2char. Changes: pg_wchar and wchar_t
  could have different size, so char2wchar doesn't call
  pg_mb2wchar_with_len to prevent out-of-bound memory bug.  Make
  char2wchar/wchar2char symmetric, now they should not be called with
  C-locale because mbstowcs/wcstombs oftenly doesn't work correct with
  C-locale.  Text parser uses pg_mb2wchar_with_len directly in case of
  C-locale and multibyte encoding Per bug report by Hiroshi Inoue and
  following discussion.  Backpatch up to 8.2 when multi-byte support
  was implemented in tsearch.

Tom Lane committed:

- When we are in error recursion trouble, arrange to suppress
  translation and encoding conversion of any elog/ereport message
  being sent to the frontend.  This generalizes a patch that I put in
  last October, which suppressed translation of only specific messages
  known to be associated with recursive can't-translate-the-message
  behavior.  As shown in bug #4680, we need a more general answer in
  order to have some hope of coping with broken encoding conversion
  setups.  This approach seems a good deal less klugy anyway.  Patch
  in all supported branches.

- In pgsql/src/backend/utils/error/elog.c, ooops ... fix some
  confusion between gettext() and _() in my previous patch.  This has
  moved around in past releases, so just copying-and-pasting from HEAD
  didn't work as intended.

- In pgsql/src/backend/utils/adt/float.c, put back our old workaround
  for machines that declare cbrt() in math.h but fail to provide the
  function itself.  Not sure how we escaped testing anything later
  than 7.3 on such cases, but they still exist, as per Andr?
  Volpato's report about AIX 5.3.

- Fix column privilege checking for cases where parent and child have
  different attribute numbering.  Also, a parent whole-row reference
  should not require select privilege on child columns that aren't
  inherited from the parent.  Problem diagnosed by KaiGai Kohei,
  though this isn't exactly his patch.

- In pgsql/doc/src/sgml/backup.sgml, remove documentation of
  log_restartpoints setting, which is obsoleted in favor of
  log_checkpoints.  Fujii Masao

- Teach the planner to support index access methods that only
  implement amgettuple or only implement amgetbitmap, instead of the
  former assumption that every AM supports both APIs.  Extracted with
  minor editorialization from Teodor Sigaev's fast-GIN-insert patch;
  whatever becomes of that, this seems like a simple and reasonable
  generalization of the index AM interface spec.

Heikki Linnakangas committed:

- In pgsql/src/include/storage/lwlock.h, remove the placeholder
  LWLockId in place of the removed FreeSpaceLock.  As pointed out by
  ITAGAKI Takahiro, we split SInvalLock into two in 8.4, so to keep
  the numbers of the rest of the locks unchanged from 8.3, we don't
  need a placeholder.

- In pgsql/src/backend/postmaster/postmaster.c, fix copy-pasto in the
  patch to allow background writer to run during recovery: if
  background writer or pgstat process dies during recovery (or any
  other child process, but those two are the only ones running), send
  SIGQUIT to the startup process using correct pid.

- In pgsql/src/backend/commands/functioncmds.c, add some sanity checks
  to CREATE CAST ... WITHOUT FUNCTION.  Disallow composite, enum and
  array types, as those are surely not binary-compatible with anything
  else because of the embedded OIDs.  Inspired by bug report by Oleg

- Reload config file in startup process on SIGHUP.  Fujii Masao.

- Add MUST (Mauritius Island Summer Time) to the list of known
  abbreviations.  Mauritius began using DST in the summer 2008-2009;
  the Olson library has been updated already.  Xavier Bugaud

Magnus Hagander committed:

- In pgsql/src/backend/libpq/hba.c, log a warning instead of shutting
  down the system if we can't load pg_hba.conf on reload (for example
  due to a permission error).  Selena Deckelmann.

- In pgsql/src/backend/libpq/hba.c, change hba load failure message to
  LOG instead of WARNING.  Per comment from Tom Lane.

- In pgsql/src/backend/libpq/hba.c, make pg_hba parsing report all
  errors in the file before aborting the load, instead of just
  reporting the first one.  Selena Deckelmann

- In pgsql/src/tools/msvc/, fix MSVC build for new ways to
  pull in keywords.c and kwlookup.c.

Peter Eisentraut committed:

- In pgsql/src/backend/storage/file/fd.c, don't actively violate the
  system limit of maximum open files (RLIMIT_NOFILE).  This avoids
  irritating kernel logs (if system overstep violations are enabled)
  and also the grsecurity alert when starting PostgreSQL.  Original
  patch by Jacek Drobiecki.

- Add new SQL:2008 error codes for invalid LIMIT and OFFSET values.
  Remove unused nonstandard error code that was perhaps intended for
  this but never used.

- Provide some proper minimal documentation for the pg_dump(all)
  --binary-upgrade option.  We don't want to commit to what it does,
  but hiding it will only cause confusion.

- In pgsql/src/backend/parser/scan.l, clarify to the translator that
  yyerror() deals with the translation of "syntax error", not the
  literal string.  I was previously confused on this matter, but I
  have now verified that everything is translated properly.

- In pgsql/src/bin/pg_dump/pg_backup_archiver.c, message tuning.

Alvaro Herrera committed:

- Separate the key word list that lived in keywords.c into a new
  header file kwlist.h, to avoid having to link the backend object
  file into other programs like pg_dump.  We can now simply symlink a
  single source file from the backend (kwlookup.c, containing the
  shared routine ScanKeywordLookup) and compile it locally, which is a
  lot cleaner.

- In pgsql/src/bin/scripts/Makefile, avoid MSVC breakage caused by my
  previous commit by not using a variable in the src/bin/scripts

- On Windows, call bind_textdomain_codeset on domains other than the
  default one, too, so that the codeset is properly mapped on the
  newly added PL domains.

- Add comments about kwlookup.c expectations.

- pg_bind_textdomain_codeset must exist only on ENABLE_NLS.

- Revert pg_bind_textdomain_codeset to a existant-but-empty function
  when ENABLE_NLS is not defined, for better compatibility of the
  backend with modules compiled the other way.  Per note from Tom Lane
  after my previous commit.

== Abgelehnte Patches (bis jetzt) ==

No one was disappointed this week :-)

== Eingesandte Patches ==

Bryce Cutt sent in another revision of the patch to improve
the performance of multi-batchHash join for skewed data sets.

KaiGai Kohei sent in another revision of his SE-PostgreSQL patches.

Fujii Masao sent in another revision of his PITR performance
improvement patch.

Josh Tolley sent in a back-patch for 8.3 intended to duplicate Selena
Deckelmann's, "Issue WARNING instead of FATAL when reload can't read
pg_hba.conf" behavior applied to HEAD this week.

Heikki Linnakangas sent in a patch for pg_standby which changes its
behavior on receiving a SIGQUIT, to immediately killing pg_standby,
triggering the FATAL death of the startup process, too.

Fujii Masao sent in another revision of the synchronous replication

Teodor Sigaev sent in another revision of the fast insert GIN patch.

Teodor Sigaev sent in another revision of the B-Tree emulation for GIN

Pavel Stehule sent in another revision of the mixed named notation in

Robert Lor sent in a two revisions of a patch to add Theo
Schlossnagle's Dtrace probes.

Zdenek Kotala sent in a patch to fix regression tests for the Czech

Andrew Dunstan sent in an optimization for parallel pg_restore.

Gregory Stark sent in a patch to head off a potential conflict between
indexes created concurrently and HOT.

Selena Deckelmann sent in a patch to have ParseConfigFile report all
parsing errors, then bail.


No Trackbacks


Display comments as Linear | Threaded

No comments

Add Comment

Enclosing asterisks marks text as bold (*word*), underscore are made via _word_.
E-Mail addresses will not be displayed and will only be used for E-Mail notifications.
To leave a comment you must approve it via e-mail, which will be sent to your address after submission.
Form options