Skip to content

Wöchentlicher PostgreSQL Newsletter - 16. Mai 2010


Der Originalartikel befindet sich unter:

http://www.postgresql.org/community/weeklynews/pwn20100516



== Wöchentlicher PostgreSQL Newsletter - 16. Mai 2010 ==

Bugfix Updates für 8.4.4, 8.3.11, 8.2.17, 8.1.21, 8.0.25 und 7.4.29
werden bald erscheinen. Fertigmachen zum Upgrade!

Die Wahlen für das PostgreSQL EU Board finden bis zum 23. Mai 2010
statt.
https://www.postgresql.eu/elections/

Michael Renner wird auf der Netways OSDC in Nürnberg, Deutschland,
am 23./24. Juni 2010 einen Vortrag halten.
http://www.netways.de/osdc/y2010/programm/

Der PDXPUG Day findet am 18. Juli 2010 im Oregon Convention
Center in Portland, OR., statt. Mehr Informationen unter:
http://wiki.postgresql.org/wiki/PDXPUGDay2010

== PostgreSQL Produkt Neuigkeiten ==

tail_n_mail 1.10.3, ein Perl Programm zum Suchen in Logdateien und
Mailen der Ergebnisse, ist erschienen.
http://bucardo.org/wiki/Tail_n_mail

neo_my2pg 0.8, ein Werkzeug zum Konvertieren von MySQL,
ist erschienen.
http://pgfoundry.org/projects/my2postgres/

pgAdmin 1.10.3, ein GUI-basierendes Management
Werkzeug für PostgreSQL, ist erschienen.
http://www.pgadmin.org/download/

pgbouncer 1.3.3, ein leichtgewichtiger Connectionpooler für
PostgreSQL, ist erschienen.
http://pgfoundry.org/projects/pgbouncer/

pglesslog 1.4.1, ein Set von Werkzeugen um die Größe von WALs
zu reduzieren, ist erschienen.
http://pglesslog.projects.postgresql.org/

psycopg2 2.2.0, ein in Python geschriebener Connector für
PostgreSQL, ist erschienen.
http://initd.org/psycopg/

== PostgreSQL 9.0 Feature der Woche ==

pg_ctl hat jetzt eine initdb Option.

== PostgreSQL Jobs im Mai ==

http://archives.postgresql.org/pgsql-jobs/2010-05/threads.php

== PostgreSQL Lokal ==

PgCon 2010 findet vom 20.-21. Mai 2010 in Ottawa statt mit Tutorials
vorher am 18. und 19.
http://www.pgcon.org/2010/registration.php

Der Prager PostgreSQL Testtag findet am 21. Mai 2010 in Prag,
Tschechische Repuplik, statt.
http://www.nic.cz/page/747/test-the-new-beta-version-of-postgresql-in-the-cz.nic-laboratories/

Registrierung für Open Source Bridge ist offen und Vorträge sind
veröffentlicht. Vom 1. bis 4. Juni in Portland, Oregon.
http://opensourcebridge.org/events/2010/proposals/

Michael Renner wird auf der AMOOCON, welche vom 4. bis 6. Juni 2010
stattfindet, sprechen.
http://www.amoocon.de/speakers/214

Das Southeast Linuxfest 2010 findet in Spartanburg, SC, USA am
12. und 13. Juni statt. Die PostgreSQL-Gurus Joshua Drake und
Andrew Dunstan werden einige PostgreSQL-relevante Vorträge halten,
die du gesehen haben musst.
http://southeastlinuxfest.org/

OSCON wird in Portland, Oregon vom 19. bis 23. Juli 2010 stattfinden.
http://www.oscon.com/oscon2010

== PostgreSQL in den News ==

Planet PostgreSQL: http://planet.postgresql.org/

Dieser wöchentliche PostgreSQL Newsletter wurde erstellt von David
Fetter und Devrim GUNDUZ.

Sende Neuigkeiten und Ankündigungen bis Sonntag, 15 Uhr Pazifischer
Zeit. Bitte sende englische Beiträge an david@fetter.org, deutsche an
pwn@pgug.de, italienische an pwn@itpug.org.



== Angewandte Patches ==

Tom Lane committed:

- In pgsql/src/backend/optimizer/plan/planagg.c, when adding a "target
  IS NOT NULL" indexqual to the plan for an index-optimized MIN or
  MAX, we must take care to insert the added qual in a legal place
  among the existing indexquals, if any.  The btree index AM requires
  the quals to appear in index-column order.  We didn't have to worry
  about this before because "target IS NOT NULL" was just treated as a
  plain scan filter condition; but as of 9.0 it can be an index qual
  and then it has to follow the rule.  Per report from Ian Barwick.

- In pgsql/src/backend/optimizer/prep/prepunion.c, fix incorrect patch
  that removed permission checks on inheritance child tables --- the
  parent table no longer got checked, either.  Per bug #5458 from
  Takahiro Itagaki.

- In pgsql/src/backend/postmaster/pgarch.c, cause the archiver process
  to adopt new postgresql.conf settings (particularly archive_command)
  as soon as possible, namely just before issuing a new call of
  archive_command, even when there is a backlog of files to be
  archived.  The original coding would only absorb new settings after
  clearing the backlog and returning to the outer loop.  Per
  discussion.  Back-patch to 8.3.  The logic in prior versions is a
  bit different and it doesn't seem worth taking any risks of breaking
  it.

- Add PKST to the default set of timezone abbreviations.  Per
  discussion, if we have PKT in there then PKST should be too.  Also,
  fix mistaken claim that these abbrevs are not known to zic.

- Update time zone data files to tzdata release 2010j: DST law changes
  in Argentina, Australian Antarctic, Bangladesh, Mexico, Morocco,
  Pakistan, Palestine, Russia, Syria, Tunisia.  Historical corrections
  for Taiwan.

- Clean up unnecessary unportability and compiler warnings by removing
  the cmp parameter for pg_scandir().  The code failed to support this
  anyway for Sun/Windows, so pretending we could accept a parameter
  other than NULL was just asking for trouble.

- Hook pg_upgrade into the contrib makefile structure so it gets built
  on the buildfarm.

- Preliminary release notes for releases 8.4.4, 8.3.11, 8.2.17,
  8.1.21, 8.0.25, 7.4.29.

- In pgsql/contrib/pg_upgrade_support/Makefile, fix wrong subdir.  Per
  buildfarm.

- Use "TOAST table" in place of the vague, not-used-elsewhere phrase
  "supplementary storage table".

- In pgsql/doc/src/sgml/rules.sgml, use an entity instead of non-ASCII
  letter.  Thom Brown.

- Fix up lame idea of not using autoconf to determine if platform has
  scandir().  Should fix buildfarm failures.

- In pgsql/contrib/pg_upgrade/file.c, spell _darwin_ correctly, to
  suppress compiler warning on OSX.

- In pg_upgrade, get rid of unsafe sprintf and snprintf usages.  Per
  compiler warnings.

- In pgsql/contrib/pg_upgrade/controldata.c, remove another unsafe use
  of random data as a format string.

- In pgsql/contrib/pg_upgrade/info.c, fix another obvious bug pointed
  out by a compiler warning.

- In pgsql/src/include/port.h, spell __NetBSD__ the same way
  everywhere.  Per Giles Lean.

- In pgsql/doc/src/sgml/ref/pg_restore.sgml, improve documentation of
  pg_restore's -l and -L switches to point out their interactions with
  filtering switches, such as -n and -t.  Per a complaint from Russell
  Smith.

- Ensure that pg_restore -l will output DATABASE entries whether or
  not -C is specified.  Per bug report from Russell Smith and ensuing
  discussion.  Since this is a corner case behavioral change, I'm
  going to be conservative and not back-patch it.  In passing, also
  rename the RestoreOptions field for the -C switch to something less
  generic than "create".

ITAGAKI Takahiro committed:

- In pgsql/src/backend/catalog/pg_proc.c, set per-function GUC
  settings during validating the function.  Now validators work
  properly even when the settings contain parameters that affect
  behavior of the function, like search_path.  Reported by Erwin
  Brandstetter.

- In pgsql/src/backend/po/ja.po, remove CRs.

Robert Haas committed:

- In pgsql/doc/src/sgml/ref/cluster.sgml, mention related ALTER TABLE
  variants in documentation for CLUSTER.  As suggested by Andy Lester.

- In pgsql/src/include/utils/rbtree.h, change typedef for
  rb_appendator to avoid conflict with C++ reserved words.  Fixes a
  complaint from src/tools/pginclude/cpluspluscheck reported by Peter
  Eisentraut.

- In pgsql/src/backend/postmaster/postmaster.c, we now accept
  read-only connections in state PM_RECOVERY_CONSISTENT.

- Rename PM_RECOVERY_CONSISTENT and PMSIGNAL_RECOVERY_CONSISTENT.  The
  new names PM_HOT_STANDBY and PMSIGNAL_BEGIN_HOT_STANDBY more
  accurately reflect their actual function.

- In pgsql/doc/src/sgml/xfunc.sgml, fix longstanding typo in V1
  calling conventions documentation.  Erik Rijkers

- In pgsql/doc/src/sgml/xfunc.sgml, insert line breaks in two places
  in SQL functions documentation.  This avoids a formatting problem in
  the PDF output.  In the HTML output this isn't necessary, but we've
  done similar things elsewhere in the documentation so I think it's
  OK to do it here, too.  I've refrained from breaking a longish error
  message which also causes problems for the PDF output, because that
  would make the HTML output look wrong.  Erik Rijkers.

Bruce Momjian committed:

- Add pg_upgrade to /contrib;  will be in 9.0 beta2.  Add
  documentation.  Supports migration from PG 8.3 and 8.4.

- Add pgsql/contrib/pg_upgrade/TODO file to CVS.

- Add pgsql/contrib/pg_upgrade/IMPLEMENTATION file to CVS.

- In pgsql/contrib/pg_upgrade/info.c, small formatting adjustment.

- Add PGFILEDESC description to Makefiles for all /contrib
  executables.  Add PGAPPICON to all executable makefiles.

- Move pg_upgrade TODO to TODO wiki.

- In pgsql/src/backend/Makefile, remove Makefile PGFILEDESC tag that
  the postmaster is an executable.

- In pgsql/src/backend/Makefile, update comment about why postmaster
  doesn't get an icon.

- Move pg_upgrade shared library out into its own /contrib directory
  (pg_upgrade_support).

- In pgsql/doc/src/sgml/release-9.0.sgml, update release notes to
  current.

- In pgsql/doc/src/sgml/release-9.0.sgml, fix HISTORY.html build using
  </link>, not </>.

- In pgsql/doc/src/sgml/pgupgrade.sgml, comment out EnterpriseDB
  Advanced Server mention in SGML docs.

- Remove all mentions of EnterpriseDB Advanced Server from pg_upgrade;
  EDB must maintain their own patch set for this.

- Use NAMEDATALEN instead of local define.  Modify path separators for
  Win32.  Per ideas from Takahiro Itagaki.

- In pgsql/contrib/pg_upgrade/file.c, change bsdi define to __bsdi__.

- Use __bsdi__ consistently.

- In pgsql/contrib/pg_upgrade/file.c, adjust pg_upgrade BSD defines
  for scandir(), per report from Giles Lean.

Peter Eisentraut committed:

- Update config.guess and config.sub.

- Translation update

- Fix vpath installation from distribution tarball (bug #5447)

- In pgsql/src/nls-global.mk, avoid error from mkdir if no languages
  are to be installed mkinstalldirs used to handle no arguments, but
  mkdir doesn't.  Also remove the .SILENT setting, that was previously
  removed from Makefile.global as well.

- Translation update

Simon Riggs committed:

- In pgsql/src/backend/tcop/postgres.c, give most recovery conflict
  errors a retryable error code.  From recent requests and discussions
  with Yeb Havinga and Kevin Grittner.

- Cleanup initialization of Hot Standby. Clarify working with
  reanalysis of requirements and documentation on
  LogStandbySnapshot().  Fixes two minor bugs reported by Tom Lane
  that would lead to an incorrect snapshot after transaction
  wraparound.  Also fix two other problems discovered that would give
  incorrect snapshots in certain cases.  ProcArrayApplyRecoveryInfo()
  substantially rewritten.  Some minor refactoring of
  xact_redo_apply() and ExpireTreeKnownAssignedTransactionIds().

- In pgsql/src/backend/access/transam/xact.c, ensure that top level
  aborts call XLogSetAsyncCommit().  Not doing so simply leads to data
  waiting in wal_buffers which then causes later commits to
  potentially do emergency writes and for all forms of replication to
  be potentially delayed without need or benefit.  Issue pointed out
  exactly by Fujii Masao, following bug report by Robert Haas on a
  separate though related topic.

- Add many new Asserts in code and fix simple bug that slipped through
  without them, related to previous commit.  Report by Bruce Momjian.

- In pgsql/src/backend/access/transam/xlog.c, fix bug in processing of
  checkpoint time for max_standby_delay.  Latest log time was
  incorrectly set, typically leading to dates in the past, which would
  cause more cancellations in Hot Standby on a quiet server.

Andrew Dunstan committed:

- In pgsql/src/tools/msvc/Mkvcbuild.pm, add missing library and
  include support for pg_upgrade to MSVC build system.

Magnus Hagander committed:

- Fix some spelling errors.  Thom Brown.

- In pgsql/src/tools/msvc/Mkvcbuild.pm, properly support multi-line
  entires (such as OBJS=) when building PROGRAM, not just MODULE, in
  contrib.

- Assorted fixes to make pg_upgrade build on MSVC.  1. There is no
  chmod() on Windows.  2. Must always use the 3-parameter version of
  open().  3. There is no dynloader.h - but it also appears unnecessary
  on all platforms.  4. Don't include shlobj.h because it causes compile
  errors, and from what I can see it's not actually used.  This may
  need to be added back for mingw and/or cygwin in the worst case.

Marc Fournier committed:

- Tag 8.4.4, 8.3.11, 8.2.17, 8.1.21, 8.0.25, 7.4.29.

Heikki Linnakangas committed:

- In pgsql/contrib/pg_standby/pg_standby.c, fix typos in comments,
  spotted by Josh Kupershmidt.

== Abgelehnte Patches (bis jetzt) ==

Greg Sabino Mullane's patch to change pg_dump not to use CREATE OR
REPLACE LANGUAGE when talking to pre-9.0 databases.  There never was a
guarantee that the output of pg_dump would be backward-compatible.

== Eingesandte Patches ==

ITAGAKI Takahiro sent in a patch to change the search_path check for
function bodies to the current search_path, where they had been using
the one at create time.

Mike Fowler sent in a patch to add a function xpath_exists() for 9.1.

Alexander Korotkov sent in two revisions of a patch to make the
levenshtein distance function in contrib/fuzzystrmatch work with
multibyte character sets in 9.1.

Simon Riggs sent in a patch to change the startup code for Hot
Standby.

Stephen Frost sent in a patch to add a handler for SIGCHLD to psql.
This is for 9.1.

Simon Riggs sent in a patch to add a keepalive message to ensure
max_standby_delay is useful.  Heikki Linnakangas responded with one
which rearranges the walsender loops slightly so that XLogSend() now
only sends up to MAX_SEND_SIZE bytes (== XLOG_SEG_SIZE / 2) in one
round and returns to the main loop after that even if there's unsent
WAL, and the main loop no longer sleeps if there's unsent WAL.  Simon
Riggs sent in another with wait processing cleanup.

Joachim Wieland sent in a patch to fix some locking behavior for 9.1
in the case of SERIALIZABLE transactions.

Robert Haas sent in a patch to implement a fix for smart shutdown in
recovery.

  • Twitter
  • Bookmark Wöchentlicher PostgreSQL Newsletter - 16. Mai 2010 at del.icio.us
  • Facebook
  • Google Bookmarks
  • FriendFeed
  • Digg Wöchentlicher PostgreSQL Newsletter - 16. Mai 2010
  • Bloglines Wöchentlicher PostgreSQL Newsletter - 16. Mai 2010
  • Technorati Wöchentlicher PostgreSQL Newsletter - 16. Mai 2010
  • Fark this: Wöchentlicher PostgreSQL Newsletter - 16. Mai 2010
  • Bookmark Wöchentlicher PostgreSQL Newsletter - 16. Mai 2010 at YahooMyWeb
  • Bookmark Wöchentlicher PostgreSQL Newsletter - 16. Mai 2010 at Furl.net
  • Bookmark Wöchentlicher PostgreSQL Newsletter - 16. Mai 2010 at reddit.com
  • Bookmark Wöchentlicher PostgreSQL Newsletter - 16. Mai 2010 at blinklist.com
  • Bookmark Wöchentlicher PostgreSQL Newsletter - 16. Mai 2010 at Spurl.net
  • Bookmark Wöchentlicher PostgreSQL Newsletter - 16. Mai 2010 at Simpy.com
  • Bookmark Wöchentlicher PostgreSQL Newsletter - 16. Mai 2010 at blogmarks
  • Bookmark Wöchentlicher PostgreSQL Newsletter - 16. Mai 2010 with wists
  • wong it!
  • Bookmark using any bookmark manager!
  • Stumble It!
  • Identi.ca

Trackbacks

No Trackbacks

Comments

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