Menu
▾
▴
postgres-xc-committers — Announces PG-XC repository commits
You can subscribe to this list here.
| 2010 |
Jan
|
Feb
|
Mar
|
Apr
(4) |
May
(28) |
Jun
(12) |
Jul
(11) |
Aug
(12) |
Sep
(5) |
Oct
(19) |
Nov
(14) |
Dec
(12) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2011 |
Jan
(18) |
Feb
(30) |
Mar
(115) |
Apr
(89) |
May
(50) |
Jun
(44) |
Jul
(22) |
Aug
(13) |
Sep
(11) |
Oct
(30) |
Nov
(28) |
Dec
(39) |
| 2012 |
Jan
(38) |
Feb
(18) |
Mar
(43) |
Apr
(91) |
May
(108) |
Jun
(46) |
Jul
(37) |
Aug
(44) |
Sep
(33) |
Oct
(29) |
Nov
(36) |
Dec
(15) |
| 2013 |
Jan
(35) |
Feb
(611) |
Mar
(5) |
Apr
(55) |
May
(30) |
Jun
(28) |
Jul
(458) |
Aug
(34) |
Sep
(9) |
Oct
(39) |
Nov
(22) |
Dec
(32) |
| 2014 |
Jan
(16) |
Feb
(16) |
Mar
(42) |
Apr
(179) |
May
(7) |
Jun
(6) |
Jul
(9) |
Aug
|
Sep
(4) |
Oct
|
Nov
(3) |
Dec
|
| 2015 |
Jan
|
Feb
|
Mar
|
Apr
(2) |
May
(4) |
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
| S | M | T | W | T | F | S |
|---|---|---|---|---|---|---|
|
|
|
|
|
|
1
|
2
(2) |
|
3
(1) |
4
|
5
(2) |
6
|
7
(9) |
8
(4) |
9
|
|
10
|
11
|
12
(4) |
13
(1) |
14
(1) |
15
(4) |
16
|
|
17
|
18
(2) |
19
(2) |
20
|
21
(5) |
22
|
23
|
|
24
|
25
|
26
(2) |
27
|
28
(4) |
29
(3) |
30
|
|
From: Michael P. <mic...@us...> - 2012-06-12 04:14:04
|
Project "Postgres-XC".
The branch, REL1_0_STABLE has been updated
via 636412d52b022bf4fde53bb7d2439a7c61fbc7eb (commit)
from 7435895479456623f6645add06d46afb5e9e5402 (commit)
- Log -----------------------------------------------------------------
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=636412d52b022bf4fde53bb7d2439a7c61fbc7eb
commit 139323f9ffe40e2e448ddeeff11559669e0d39f6
Author: Michael Paquier <mi...@ot...>
Date: Tue Jun 12 13:15:43 2012 +0900
Correct grammar of CREATE TABLE and CLEAN CONNECTION in docs
Some brackets were used on mandatory arguments of some clauses.
M doc-xc/src/sgml/ref/clean_connection.sgmlin
M doc-xc/src/sgml/ref/create_table.sgmlin
M src/backend/parser/gram.y
-----------------------------------------------------------------------
Summary of changes:
doc-xc/src/sgml/ref/clean_connection.sgmlin | 2 +-
doc-xc/src/sgml/ref/create_table.sgmlin | 4 ++--
src/backend/parser/gram.y | 4 ++--
3 files changed, 5 insertions(+), 5 deletions(-)
hooks/post-receive
--
Postgres-XC
|
|
From: Michael P. <mic...@us...> - 2012-06-12 00:41:04
|
Project "Postgres-XC".
The branch, master has been updated
via 9a7d611371b8ba95d250844148869ccc487b2184 (commit)
from 32f458a07a81d23d834cb8845d58961c23919eff (commit)
- Log -----------------------------------------------------------------
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=9a7d611371b8ba95d250844148869ccc487b2184
commit 9a7d611371b8ba95d250844148869ccc487b2184
Author: Michael Paquier <mi...@ot...>
Date: Tue Jun 12 09:40:36 2012 +0900
Make data ordering consistent in query for test join.sql
Regression test join failed on rare occasions depending on the cluster
configuration.
M src/test/regress/expected/join.out
M src/test/regress/expected/join_1.out
M src/test/regress/sql/join.sql
-----------------------------------------------------------------------
Summary of changes:
src/test/regress/expected/join.out | 4 ++--
src/test/regress/expected/join_1.out | 33 ++++++++++++++++++---------------
src/test/regress/sql/join.sql | 4 ++--
3 files changed, 22 insertions(+), 19 deletions(-)
hooks/post-receive
--
Postgres-XC
|
|
From: Michael P. <mic...@us...> - 2012-06-12 00:41:01
|
Project "Postgres-XC".
The branch, REL1_0_STABLE has been updated
via 7435895479456623f6645add06d46afb5e9e5402 (commit)
from d141ce2dc7811dff101e52094eb18acf719c449d (commit)
- Log -----------------------------------------------------------------
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=7435895479456623f6645add06d46afb5e9e5402
commit 9a7d611371b8ba95d250844148869ccc487b2184
Author: Michael Paquier <mi...@ot...>
Date: Tue Jun 12 09:40:36 2012 +0900
Make data ordering consistent in query for test join.sql
Regression test join failed on rare occasions depending on the cluster
configuration.
M src/test/regress/expected/join.out
M src/test/regress/expected/join_1.out
M src/test/regress/sql/join.sql
-----------------------------------------------------------------------
Summary of changes:
src/test/regress/expected/join.out | 4 ++--
src/test/regress/expected/join_1.out | 33 ++++++++++++++++++---------------
src/test/regress/sql/join.sql | 4 ++--
3 files changed, 22 insertions(+), 19 deletions(-)
hooks/post-receive
--
Postgres-XC
|
|
From: Michael P. <mic...@us...> - 2012-06-08 00:36:29
|
Project "Postgres-XC".
The branch, master has been updated
via 32f458a07a81d23d834cb8845d58961c23919eff (commit)
from 4ce5ac8c5a91012d139449b8f97b94b5695ae561 (commit)
- Log -----------------------------------------------------------------
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=32f458a07a81d23d834cb8845d58961c23919eff
commit 32f458a07a81d23d834cb8845d58961c23919eff
Author: Michael Paquier <mi...@ot...>
Date: Fri Jun 8 09:37:29 2012 +0900
Replace node type by correct terms in CREATE/ALTER NODE and short install manual
Terms in command CREATE/ALTER NODE need to be lower case.
M doc-xc/src/sgml/installation.sgmlin
M doc-xc/src/sgml/ref/alter_node.sgmlin
M doc-xc/src/sgml/ref/create_node.sgmlin
-----------------------------------------------------------------------
Summary of changes:
doc-xc/src/sgml/installation.sgmlin | 20 ++++++++++----------
doc-xc/src/sgml/ref/alter_node.sgmlin | 2 +-
doc-xc/src/sgml/ref/create_node.sgmlin | 6 +++---
3 files changed, 14 insertions(+), 14 deletions(-)
hooks/post-receive
--
Postgres-XC
|
|
From: Michael P. <mic...@us...> - 2012-06-08 00:36:26
|
Project "Postgres-XC".
The branch, REL1_0_STABLE has been updated
via d141ce2dc7811dff101e52094eb18acf719c449d (commit)
from beee077699b85fd51c452a7d5d6b2eeead216cfb (commit)
- Log -----------------------------------------------------------------
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=d141ce2dc7811dff101e52094eb18acf719c449d
commit 32f458a07a81d23d834cb8845d58961c23919eff
Author: Michael Paquier <mi...@ot...>
Date: Fri Jun 8 09:37:29 2012 +0900
Replace node type by correct terms in CREATE/ALTER NODE and short install manual
Terms in command CREATE/ALTER NODE need to be lower case.
M doc-xc/src/sgml/installation.sgmlin
M doc-xc/src/sgml/ref/alter_node.sgmlin
M doc-xc/src/sgml/ref/create_node.sgmlin
-----------------------------------------------------------------------
Summary of changes:
doc-xc/src/sgml/installation.sgmlin | 20 ++++++++++----------
doc-xc/src/sgml/ref/alter_node.sgmlin | 2 +-
doc-xc/src/sgml/ref/create_node.sgmlin | 6 +++---
3 files changed, 14 insertions(+), 14 deletions(-)
hooks/post-receive
--
Postgres-XC
|
|
From: Michael P. <mic...@us...> - 2012-06-08 00:16:11
|
Project "Postgres-XC".
The branch, master has been updated
via 4ce5ac8c5a91012d139449b8f97b94b5695ae561 (commit)
from e4b96326668a182568f6b74ef83b64fc8c95d885 (commit)
- Log -----------------------------------------------------------------
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=4ce5ac8c5a91012d139449b8f97b94b5695ae561
commit 4ce5ac8c5a91012d139449b8f97b94b5695ae561
Author: Michael Paquier <mi...@ot...>
Date: Fri Jun 8 09:18:08 2012 +0900
Correction in installation doc on commands with a prompt tag missing
Report and patch by Michael Meskes
M doc-xc/src/sgml/runtime.sgmlin
-----------------------------------------------------------------------
Summary of changes:
doc-xc/src/sgml/runtime.sgmlin | 12 ++++++------
1 files changed, 6 insertions(+), 6 deletions(-)
hooks/post-receive
--
Postgres-XC
|
|
From: Michael P. <mic...@us...> - 2012-06-08 00:16:06
|
Project "Postgres-XC".
The branch, REL1_0_STABLE has been updated
via beee077699b85fd51c452a7d5d6b2eeead216cfb (commit)
from 4af426e41e35322ace0797bf4ddb0c3efdc22fd7 (commit)
- Log -----------------------------------------------------------------
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=beee077699b85fd51c452a7d5d6b2eeead216cfb
commit 4ce5ac8c5a91012d139449b8f97b94b5695ae561
Author: Michael Paquier <mi...@ot...>
Date: Fri Jun 8 09:18:08 2012 +0900
Correction in installation doc on commands with a prompt tag missing
Report and patch by Michael Meskes
M doc-xc/src/sgml/runtime.sgmlin
-----------------------------------------------------------------------
Summary of changes:
doc-xc/src/sgml/runtime.sgmlin | 12 ++++++------
1 files changed, 6 insertions(+), 6 deletions(-)
hooks/post-receive
--
Postgres-XC
|
|
From: Michael P. <mic...@us...> - 2012-06-07 06:37:09
|
Project "website".
The branch, master has been updated
via ec56e2c9f10e7e434b3a16e17260353f8c0e1ce8 (commit)
from 60bd2e69d0f7eaca2b3c514e62656046d8feb5da (commit)
- Log -----------------------------------------------------------------
commit ec56e2c9f10e7e434b3a16e17260353f8c0e1ce8
Author: Michael Paquier <mi...@ot...>
Date: Thu Jun 7 15:41:21 2012 +0900
Update website with 1.0.0 information
diff --git a/download.html b/download.html
index bb9d80f..6cf3460 100755
--- a/download.html
+++ b/download.html
@@ -15,15 +15,11 @@
<link rel="stylesheet" type="text/css" href="pgcx_1.css">
</head>
<body bgcolor=#FFFFFF>
-
-<h2>
-Downloading Postgres-XC
-</h2>
-
-<h3>
-List of Release Materials
-</h3>
-
+
+<h2>Downloading Postgres-XC</h2>
+
+<h3>List of Release Materials</h3>
+
<p>
The current release includes the following materials, which can be downloaded directly from <br>
<a href="https://sourceforge.net/projects/postgres-xc/files/" target="_blank">the project download page.</a>
@@ -32,61 +28,25 @@ The current release includes the following materials, which can be downloaded di
Since Postgres-XC version 1.0, tarballs include HTML and man documentation.
</p>
-<!-- Documents of version 1.0beta2 -->
-<h4>Version 1.0beta2</h4>
-<p>
-<ul>
-<!-- tarball of 1.0beta2, main download-->
-<li>
-<code>pgxc_v1.0beta2.tar.gz</code>: <br>
-Latest version of Postgres-XC available.<br>
-Please note that Postgres-XC documentation is not included in this file.
-⇒
-<a href="https://sourceforge.net/projects/postgres-xc/files/Version_1.0beta2/pgxc_v1.0beta2.tar.gz/download" target="_blank">(download)</a>
-</li>
-
-<!-- License -->
-<li>
-<code>COPYING</code>: <br>
-License description. Postgres-XC is distributed under the PostgreSQL license.
-⇒
-<a href="https://sourceforge.net/projects/postgres-xc/files/Version_1.0beta2/COPYING/download" target="_blank">
-(download)
-</a>
-</li>
-
-<!-- Files -->
-<li>
-<code>FILES</code>: <br>
-Description of files included in Postgres-XC 1.0beta2 release.
-⇒
-<a href="https://sourceforge.net/projects/postgres-xc/files/Version_1.0beta2/FILES/download" target="_blank">
-(download)
-</a>
-</li>
-
-</ul>
-</p>
-
-<!-- Documents of version 1.0beta1 -->
-<h4>Version 1.0beta1</h4>
+<!-- Documents of version 1.0 -->
+<h4>Version 1.0</h4>
<p>
<ul>
-<!-- tarball of 1.0beta1, main download-->
+<!-- tarball of 1.0, main download-->
<li>
-<code>pgxc_v1.0beta1.tar.gz</code>: <br>
+<code>Postgres-XC 1.0.0 source tarball</code>: <br>
Latest version of Postgres-XC available.<br>
Please note that Postgres-XC documentation is not included in this file.
⇒
-<a href="https://sourceforge.net/projects/postgres-xc/files/Version_1.0beta1/pgxc_v1.0beta1.tar.gz/download" target="_blank">(download)</a>
+<a href="https://sourceforge.net/projects/postgres-xc/files/Version_1.0/pgxc-v1.0.0.tar.gz/download" target="_blank">(download)</a>
</li>
<!-- License -->
<li>
-<code>COPYING</code>: <br>
+<code>COPYRIGHT</code>: <br>
License description. Postgres-XC is distributed under the PostgreSQL license.
⇒
-<a href="https://sourceforge.net/projects/postgres-xc/files/Version_1.0beta1/COPYING/download" target="_blank">
+<a href="https://sourceforge.net/projects/postgres-xc/files/Version_1.0/COPYRIGHT/download" target="_blank">
(download)
</a>
</li>
@@ -94,9 +54,9 @@ License description. Postgres-XC is distributed under the PostgreSQL license.
<!-- Files -->
<li>
<code>FILES</code>: <br>
-Description of files included in Postgres-XC 1.0beta1 release.
+Description of files included in Postgres-XC 1.0 release.
⇒
-<a href="https://sourceforge.net/projects/postgres-xc/files/Version_1.0beta1/FILES/download" target="_blank">
+<a href="https://sourceforge.net/projects/postgres-xc/files/Version_1.0/FILES/download" target="_blank">
(download)
</a>
</li>
diff --git a/events.html b/events.html
index 657ce05..9645ec9 100755
--- a/events.html
+++ b/events.html
@@ -15,7 +15,6 @@
<p class="plain">
A lot of opportunities to meet the Core developers!!
<ul>
-<li>Tutorial at PGCon on May 2012 at Ottawa,Canada</li>
<li>Presentation at PGCon China in June at Beijing, China</li>
</ul>
</p>
@@ -30,13 +29,11 @@ Event title
Description of this event.
</p>
-->
-<!--
- UPDATES
--->
+<!-- UPDATES -->
<h2 class="plain">Updates</h2>
-<!-- Postgres-XC 1.0beta2 download -->
+<!-- Postgres-XC 1.0.0 download -->
<p class="plain">
-Postgres-XC 1.0beta2 is now available!! Download
+Postgres-XC 1.0.0 is now available!! Download
<a href="https://sourceforge.net/projects/postgres-xc/files/latest/download" target="_blank">
here.
</a>
diff --git a/roadmap.html b/roadmap.html
index 3f407d6..b5f7c2d 100755
--- a/roadmap.html
+++ b/roadmap.html
@@ -22,86 +22,47 @@ TODO: Should prototyping/application partner issue be included here?
Postgres-XC Roadmap
</h2>
-<!-- ==== Current Limitation ==== -->
-<h3>
-Current Limitations of Postgres-XC
-</h3>
-
-<p>
-At present, Postgres-XC provides major transaction management features
-similar to PostgreSQL, except for savepoints.
-</p>
-<p>
-As of Version 1.0beta2, Postgres-XC supports statements which can be executed
+<!-- ==== Latest features ==== -->
+<h3>Latest features of Postgres-XC</h3>
+
+<p>At present, Postgres-XC provides major transaction management features
+similar to PostgreSQL, except for savepoints.</p>
+<p>As of Version 1.0, Postgres-XC supports statements which can be executed
and support for identification of expressions non-shippable to remote nodes.<br />
This new version is centralized on stabilization and adds support for:</p>
<ul>
<li>Fast query shipping (FQS)</li>
<li>SERIAL</li>
<li>TABLESPACE</li>
+<li>EXTENSION</li>
<li>Rules</li>
<li>Utility to clean up 2PC transactions in cluster (pgxc_clean)</li>
<li>Utility for initialization of GTM (initgtm)</li>
<li>Relation-size functions</li>
<li>All 139 regressions passing</li>
<li>Support of "make world"</li>
+<li>Support for vpath installation</li>
+<li>Support of compilation on MacOS</li>
</ul>
-<p>Currently in beta1 state, the next effort to a stable version will be stabilization and bug fix.</p>
<!-- ==== Planned feature === -->
-<h3>
-Upcoming Releases and Features
-</h3>
-
-<p>
-Current plan of future releases and features are as follows:
-</p>
+<h3>Upcoming Releases and Features</h3>
-<!-- ==== For version 0.9.3 ==== -->
-<!-- <h4>
-Version 0.9.3 (Late in September, 2010)
-</h4>
+<p>Current plan of future releases and features are as follows:</p>
-<p class="inner">
-Cluster-wide installer<br>
-Cluster-wide operation utilities<br>
-Regression tests<br>
-Logical backup/restore (pg_dump, pg_restore)<br>
-Basic cross-node operation<br>
-TEMP Table<br>
-Cursor support<br>
-Extended Query Protocol (for JDBC)<br>
-Global timestamp<br>
-Driver support (ECPG, JDBC, PHP, etc.)<br>
-Forward Cursor (w/o <code>ORDER BY</code>)<br>
-subqueries<br>
-</p> -->
-
-<h4>Next release 1.0.0</h4>
-<ul>
-<li>Fast query shipping (FQS)</li>
-<li>SERIAL</li>
-<li>TABLESPACE</li>
-<li>Utility to clean up 2PC transactions in cluster (pgxc_clean)</li>
-<li>Utility for initialization of GTM (initgtm)</li>
-<li>Relation-size functions</li>
-<li>Regression stabilization and diff cleanup</li>
-<li>Rules</li>
-<li>Stable regressions</li>
-<li>Support for make world</li>
-</ul>
-
<h4>
-Next release V1.1 (October 2012)
+Next release V1.1 (May 2013)
</h4>
<p class="inner">
<ul>
+<li>Merge with PostgreSQL 9.2</li>
<li>Data redistribution</li>
-<li>Tuple relocation (distribute key update)</li>
+<li>Planner improvements</li>
<li>Online server addition and removal</li>
-<li>Cluster bootstrap</li>
<li>Triggers</li>
+<li>Cross-node operation optimization</li>
+<li>Global constraints</li>
</ul>
</p>
@@ -112,12 +73,10 @@ Beyond 1.1
<p class="inner">
<ul>
-<li>SSL connection protocol for pooler (secure your cluster!)</li>
-<li>HOT standby full support</li>
-<li>Cross-node operation optimization</li>
-<li>Savepoint</li>
+<li>Global deadlock detection</li>
+<li>Foreign Data wrapper</li>
<li>Batch, statement pushdown</li>
-<li>Global constraints</li>
+<li>SSI</li>
</ul>
</p>
-----------------------------------------------------------------------
Summary of changes:
download.html | 68 ++++++++++---------------------------------------
events.html | 9 ++----
roadmap.html | 79 +++++++++++++-------------------------------------------
3 files changed, 36 insertions(+), 120 deletions(-)
hooks/post-receive
--
website
|
|
From: Michael P. <mic...@us...> - 2012-06-07 05:57:57
|
Project "Postgres-XC".
The annotated tag, XC1_0_0_PG9_1 has been created
at 9983f48fff1352ed14a97a45976bc38a9c9e5f72 (tag)
tagging 4af426e41e35322ace0797bf4ddb0c3efdc22fd7 (commit)
replaces XC1_0_BETA2_PG9_1
tagged by Michael
on Thu Jun 7 14:59:15 2012 +0900
- Log -----------------------------------------------------------------
Postgres-XC version 1.0.0 tag, based on PostgreSQL 9.1
Abbas (2):
Fix a problem in explicitly PREPARED TRANSACTIONS.
Fix a problem in explicitly PREPARED TRANSACTIONS.
Heikki Linnakangas (2):
On GiST page split, release the locks on child pages before recursing up.
Fix bug in to_tsquery().
Joe Conway (1):
PL/pgSQL RETURN NEXT was leaking converted tuples, causing
Magnus Hagander (2):
Only attempt to show collations on servers >= 9.1.
Fix handling of pg_stat_statements.stat temporary file
Michael Paquier (39):
Change xc_copy to use dedicated regression data folder
Restructure make files of GTM build and fix VPATH build on GTM side
Fix make install for vpath installation
Fix some typo in gtm and initdb documentation of XC
Fix make world for vpath installation
Fix for initgtm for share folder when build is not installed
Ignore input and output of regression test xc_copy in GIT repository
Update GTM makefiles and copyright dates
Fix for bug 3525691: autovacuum process issue
Change Copyright to "Postgres-XC Development Group"
Suppress forgotten library link in src/gtm/path for "make clean"
Add flags in LINK.shared: -undefined suppress -flat_namespace
Simplify remote node communication at PREPARE phase
Correct reference to project websites in install documentation
Remove useless query in test xc_misc
Fix a problem with transactions staying open on remote nodes
Fix pgxc_clean crash when no options are specified
Fix CREATE EXTENSION in cluster
Keep sgml files in distribution tarball
Reinforce ALTER SEQUENCE handling on GTM for START/RESTART
Change INSERT tag for remote INSERT queries
Clean up documentation about XCM, pgxc_config and xc_watcher
Modify short installation in docs with simple Postgres-XC install
Change "Postgre-XC" by "Postgres-XC" in documentation
Clarify role of parameter xc_maintenance_mode
Correct typo in page related to application gtm in docs
Avoid index collate for recovery_target_barrier when generating documentation
Add reference to gtm.conf and gtm_proxy.conf in documentation Index
Update documentation with correct naming for Coordinator and Datanode
Correct commands in documentation for initdb, postgres and initgtm
Update code comments: use "Coordinator" and "Datanode" as conventional terms
Move pgxc_clean and pgxc_ddl to contrib modules
Block node registration for slave nodes and perform register at a stable state
Merge branch 'REL9_1_STABLE' into REL1_0_STABLE
Change mode choice option from -S to -Z
Update pg_ctl with correct bug ML and reorganize help
Correct segenv in pgxc_clean due to argument initialization
Stamp HEAD as 1.0.0.
Update release date of 1.0.0 in docs
Peter Eisentraut (2):
pg_standby: Remove tabs from string literals
Translation updates
Robert Haas (2):
Prevent loss of init fork when truncating an unlogged table.
Fix error message for COMMENT/SECURITY LABEL ON COLUMN xxx IS 'yyy'
Simon Riggs (5):
Avoid xid error from age() function when run on Hot Standby
Ensure age() returns a stable value rather than the latest value
Remove extraneous #include "storage/proc.h"
Ensure backwards compatibility for GetStableLatestTransactionId()
Avoid early reuse of btree pages, causing incorrect query results.
Tom Lane (16):
Fix Windows implementation of PGSemaphoreLock.
Fix contrib/citext's upgrade script to handle array and domain cases.
Fix DROP TABLESPACE to unlink symlink when directory is not there.
Move postmaster's RemovePgTempFiles call to a less randomly chosen place.
Ensure that seqscans check for interrupts at least once per page.
Use binary search instead of brute-force scan in findNamespace().
Fix string truncation to be multibyte-aware in text_name and bpchar_name.
Prevent synchronized scanning when systable_beginscan chooses a heapscan.
Teach AbortOutOfAnyTransaction to clean up partially-started transactions.
Fix incorrect password transformation in contrib/pgcrypto's DES crypt().
Expand the allowed range of timezone offsets to +/-15:59:59 from Greenwich.
Ignore SECURITY DEFINER and SET attributes for a PL's call handler.
Update time zone data files to tzdata release 2012c.
Revert back-branch changes in behavior of age(xid).
Update release notes for 9.1.4, 9.0.8, 8.4.12, 8.3.19.
Stamp 9.1.4.
-----------------------------------------------------------------------
hooks/post-receive
--
Postgres-XC
|
|
From: Michael P. <mic...@us...> - 2012-06-07 05:57:04
|
Project "Postgres-XC".
The branch, master has been updated
via e4b96326668a182568f6b74ef83b64fc8c95d885 (commit)
from 7053dd38c1b61456a2fcd88fa02faf09865d4206 (commit)
- Log -----------------------------------------------------------------
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=e4b96326668a182568f6b74ef83b64fc8c95d885
commit e4b96326668a182568f6b74ef83b64fc8c95d885
Author: Michael Paquier <mi...@ot...>
Date: Thu Jun 7 14:55:05 2012 +0900
Update release date of 1.0.0 in docs
M doc-xc/src/sgml/release-xc-1.0.sgmlin
-----------------------------------------------------------------------
Summary of changes:
doc-xc/src/sgml/release-xc-1.0.sgmlin | 4 +---
1 files changed, 1 insertions(+), 3 deletions(-)
hooks/post-receive
--
Postgres-XC
|
|
From: Michael P. <mic...@us...> - 2012-06-07 05:57:02
|
Project "Postgres-XC".
The branch, REL1_0_STABLE has been updated
via 4af426e41e35322ace0797bf4ddb0c3efdc22fd7 (commit)
from 244028b36d7f2ce1ecc7417c4b6437af2c168782 (commit)
- Log -----------------------------------------------------------------
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=4af426e41e35322ace0797bf4ddb0c3efdc22fd7
commit e4b96326668a182568f6b74ef83b64fc8c95d885
Author: Michael Paquier <mi...@ot...>
Date: Thu Jun 7 14:55:05 2012 +0900
Update release date of 1.0.0 in docs
M doc-xc/src/sgml/release-xc-1.0.sgmlin
-----------------------------------------------------------------------
Summary of changes:
doc-xc/src/sgml/release-xc-1.0.sgmlin | 4 +---
1 files changed, 1 insertions(+), 3 deletions(-)
hooks/post-receive
--
Postgres-XC
|
|
From: Michael P. <mic...@us...> - 2012-06-07 05:53:11
|
Project "Postgres-XC".
The annotated tag, XC1_0_0_PG9_1 has been deleted
was bfc52c9c10097916e7b883d559dd138238308fa7
-----------------------------------------------------------------------
tag XC1_0_0_PG9_1
Postgres-XC version 1.0.0 tag, based on PostgreSQL 9.1
244028b36d7f2ce1ecc7417c4b6437af2c168782 Stamp HEAD as 1.0.0.
-----------------------------------------------------------------------
hooks/post-receive
--
Postgres-XC
|
|
From: Michael P. <mic...@us...> - 2012-06-07 05:12:07
|
Project "Postgres-XC".
The annotated tag, XC1_0_0_PG9_1 has been created
at bfc52c9c10097916e7b883d559dd138238308fa7 (tag)
tagging 244028b36d7f2ce1ecc7417c4b6437af2c168782 (commit)
replaces XC1_0_BETA2_PG9_1
tagged by Michael Paquier
on Thu Jun 7 14:12:41 2012 +0900
- Log -----------------------------------------------------------------
Postgres-XC version 1.0.0 tag, based on PostgreSQL 9.1
Abbas (2):
Fix a problem in explicitly PREPARED TRANSACTIONS.
Fix a problem in explicitly PREPARED TRANSACTIONS.
Heikki Linnakangas (2):
On GiST page split, release the locks on child pages before recursing up.
Fix bug in to_tsquery().
Joe Conway (1):
PL/pgSQL RETURN NEXT was leaking converted tuples, causing
Magnus Hagander (2):
Only attempt to show collations on servers >= 9.1.
Fix handling of pg_stat_statements.stat temporary file
Michael Paquier (38):
Change xc_copy to use dedicated regression data folder
Restructure make files of GTM build and fix VPATH build on GTM side
Fix make install for vpath installation
Fix some typo in gtm and initdb documentation of XC
Fix make world for vpath installation
Fix for initgtm for share folder when build is not installed
Ignore input and output of regression test xc_copy in GIT repository
Update GTM makefiles and copyright dates
Fix for bug 3525691: autovacuum process issue
Change Copyright to "Postgres-XC Development Group"
Suppress forgotten library link in src/gtm/path for "make clean"
Add flags in LINK.shared: -undefined suppress -flat_namespace
Simplify remote node communication at PREPARE phase
Correct reference to project websites in install documentation
Remove useless query in test xc_misc
Fix a problem with transactions staying open on remote nodes
Fix pgxc_clean crash when no options are specified
Fix CREATE EXTENSION in cluster
Keep sgml files in distribution tarball
Reinforce ALTER SEQUENCE handling on GTM for START/RESTART
Change INSERT tag for remote INSERT queries
Clean up documentation about XCM, pgxc_config and xc_watcher
Modify short installation in docs with simple Postgres-XC install
Change "Postgre-XC" by "Postgres-XC" in documentation
Clarify role of parameter xc_maintenance_mode
Correct typo in page related to application gtm in docs
Avoid index collate for recovery_target_barrier when generating documentation
Add reference to gtm.conf and gtm_proxy.conf in documentation Index
Update documentation with correct naming for Coordinator and Datanode
Correct commands in documentation for initdb, postgres and initgtm
Update code comments: use "Coordinator" and "Datanode" as conventional terms
Move pgxc_clean and pgxc_ddl to contrib modules
Block node registration for slave nodes and perform register at a stable state
Merge branch 'REL9_1_STABLE' into REL1_0_STABLE
Change mode choice option from -S to -Z
Update pg_ctl with correct bug ML and reorganize help
Correct segenv in pgxc_clean due to argument initialization
Stamp HEAD as 1.0.0.
Peter Eisentraut (2):
pg_standby: Remove tabs from string literals
Translation updates
Robert Haas (2):
Prevent loss of init fork when truncating an unlogged table.
Fix error message for COMMENT/SECURITY LABEL ON COLUMN xxx IS 'yyy'
Simon Riggs (5):
Avoid xid error from age() function when run on Hot Standby
Ensure age() returns a stable value rather than the latest value
Remove extraneous #include "storage/proc.h"
Ensure backwards compatibility for GetStableLatestTransactionId()
Avoid early reuse of btree pages, causing incorrect query results.
Tom Lane (16):
Fix Windows implementation of PGSemaphoreLock.
Fix contrib/citext's upgrade script to handle array and domain cases.
Fix DROP TABLESPACE to unlink symlink when directory is not there.
Move postmaster's RemovePgTempFiles call to a less randomly chosen place.
Ensure that seqscans check for interrupts at least once per page.
Use binary search instead of brute-force scan in findNamespace().
Fix string truncation to be multibyte-aware in text_name and bpchar_name.
Prevent synchronized scanning when systable_beginscan chooses a heapscan.
Teach AbortOutOfAnyTransaction to clean up partially-started transactions.
Fix incorrect password transformation in contrib/pgcrypto's DES crypt().
Expand the allowed range of timezone offsets to +/-15:59:59 from Greenwich.
Ignore SECURITY DEFINER and SET attributes for a PL's call handler.
Update time zone data files to tzdata release 2012c.
Revert back-branch changes in behavior of age(xid).
Update release notes for 9.1.4, 9.0.8, 8.4.12, 8.3.19.
Stamp 9.1.4.
-----------------------------------------------------------------------
hooks/post-receive
--
Postgres-XC
|
|
From: Michael P. <mic...@us...> - 2012-06-07 05:11:30
|
Project "Postgres-XC".
The branch, REL1_0_STABLE has been updated
via 244028b36d7f2ce1ecc7417c4b6437af2c168782 (commit)
from e334f724fd0821b412fa24336b97aae3b3b49395 (commit)
- Log -----------------------------------------------------------------
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=244028b36d7f2ce1ecc7417c4b6437af2c168782
commit 7053dd38c1b61456a2fcd88fa02faf09865d4206
Author: Michael Paquier <mi...@ot...>
Date: Thu Jun 7 13:08:57 2012 +0900
Correct segenv in pgxc_clean due to argument initialization
Launching pgxc_clean made it crash when free'ing some fields
with commands where arguments were not all specified.
Patch by Mason Sharp, some reformat by me.
M contrib/pgxc_clean/pgxc_clean.c
-----------------------------------------------------------------------
Summary of changes:
configure | 18 +++++++++---------
configure.in | 2 +-
doc-xc/bug.template | 2 +-
doc-xc/src/sgml/Makefile | 2 +-
src/include/pg_config.h.win32 | 4 ++--
5 files changed, 14 insertions(+), 14 deletions(-)
hooks/post-receive
--
Postgres-XC
|
|
From: Michael P. <mic...@us...> - 2012-06-07 04:13:02
|
Project "Postgres-XC".
The branch, master has been updated
via 7053dd38c1b61456a2fcd88fa02faf09865d4206 (commit)
from b003afe5ea7cf83f4f877161d3d7d76edce7a114 (commit)
- Log -----------------------------------------------------------------
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=7053dd38c1b61456a2fcd88fa02faf09865d4206
commit 7053dd38c1b61456a2fcd88fa02faf09865d4206
Author: Michael Paquier <mi...@ot...>
Date: Thu Jun 7 13:08:57 2012 +0900
Correct segenv in pgxc_clean due to argument initialization
Launching pgxc_clean made it crash when free'ing some fields
with commands where arguments were not all specified.
Patch by Mason Sharp, some reformat by me.
M contrib/pgxc_clean/pgxc_clean.c
-----------------------------------------------------------------------
Summary of changes:
contrib/pgxc_clean/pgxc_clean.c | 90 ++++++++++++++++++++++-----------------
1 files changed, 51 insertions(+), 39 deletions(-)
hooks/post-receive
--
Postgres-XC
|
|
From: Michael P. <mic...@us...> - 2012-06-07 04:12:59
|
Project "Postgres-XC".
The branch, REL1_0_STABLE has been updated
via e334f724fd0821b412fa24336b97aae3b3b49395 (commit)
from 7fb396c3fe1055a33326243a384910e98f524714 (commit)
- Log -----------------------------------------------------------------
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=e334f724fd0821b412fa24336b97aae3b3b49395
commit 7053dd38c1b61456a2fcd88fa02faf09865d4206
Author: Michael Paquier <mi...@ot...>
Date: Thu Jun 7 13:08:57 2012 +0900
Correct segenv in pgxc_clean due to argument initialization
Launching pgxc_clean made it crash when free'ing some fields
with commands where arguments were not all specified.
Patch by Mason Sharp, some reformat by me.
M contrib/pgxc_clean/pgxc_clean.c
-----------------------------------------------------------------------
Summary of changes:
contrib/pgxc_clean/pgxc_clean.c | 90 ++++++++++++++++++++++-----------------
1 files changed, 51 insertions(+), 39 deletions(-)
hooks/post-receive
--
Postgres-XC
|
|
From: Michael P. <mic...@us...> - 2012-06-05 12:49:01
|
Project "Postgres-XC".
The branch, master has been updated
via b003afe5ea7cf83f4f877161d3d7d76edce7a114 (commit)
via a868afadd0c3b19be26008c6f6725ee3285deebf (commit)
from dfee5379109a757e1bfaa9fe59830e6fb05f6688 (commit)
- Log -----------------------------------------------------------------
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=b003afe5ea7cf83f4f877161d3d7d76edce7a114
commit b003afe5ea7cf83f4f877161d3d7d76edce7a114
Author: Michael Paquier <mi...@ot...>
Date: Tue Jun 5 21:45:44 2012 +0900
Update pg_ctl with correct bug ML and reorganize help
The PostgreSQL mailing list for bugs was incorrectly used.
M src/bin/pg_ctl/pg_ctl.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=a868afadd0c3b19be26008c6f6725ee3285deebf
commit b003afe5ea7cf83f4f877161d3d7d76edce7a114
Author: Michael Paquier <mi...@ot...>
Date: Tue Jun 5 21:45:44 2012 +0900
Update pg_ctl with correct bug ML and reorganize help
The PostgreSQL mailing list for bugs was incorrectly used.
M src/bin/pg_ctl/pg_ctl.c
-----------------------------------------------------------------------
Summary of changes:
doc-xc/src/sgml/ref/gtm_ctl.sgmlin | 10 ++++----
src/bin/pg_ctl/pg_ctl.c | 8 +++++-
src/gtm/gtm_ctl/gtm_ctl.c | 40 ++++++++++++++++++------------------
src/gtm/test/promote.sh | 2 +-
src/gtm/test/start_a.sh | 4 +-
src/gtm/test/start_s.sh | 4 +-
src/gtm/test/stop.sh | 4 +-
src/gtm/test2/status_a.sh | 2 +-
src/gtm/test2/status_s.sh | 2 +-
src/test/regress/pg_regress.c | 2 +-
10 files changed, 41 insertions(+), 37 deletions(-)
hooks/post-receive
--
Postgres-XC
|
|
From: Michael P. <mic...@us...> - 2012-06-05 12:48:50
|
Project "Postgres-XC".
The branch, REL1_0_STABLE has been updated
via 7fb396c3fe1055a33326243a384910e98f524714 (commit)
via a5ba4faf82c246dadb3213eedb1d271c893c21ed (commit)
from dc3d84756ca6d38dedc48e24813a9adcd9a6a241 (commit)
- Log -----------------------------------------------------------------
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=7fb396c3fe1055a33326243a384910e98f524714
commit b003afe5ea7cf83f4f877161d3d7d76edce7a114
Author: Michael Paquier <mi...@ot...>
Date: Tue Jun 5 21:45:44 2012 +0900
Update pg_ctl with correct bug ML and reorganize help
The PostgreSQL mailing list for bugs was incorrectly used.
M src/bin/pg_ctl/pg_ctl.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=a5ba4faf82c246dadb3213eedb1d271c893c21ed
commit b003afe5ea7cf83f4f877161d3d7d76edce7a114
Author: Michael Paquier <mi...@ot...>
Date: Tue Jun 5 21:45:44 2012 +0900
Update pg_ctl with correct bug ML and reorganize help
The PostgreSQL mailing list for bugs was incorrectly used.
M src/bin/pg_ctl/pg_ctl.c
-----------------------------------------------------------------------
Summary of changes:
doc-xc/src/sgml/ref/gtm_ctl.sgmlin | 10 ++++----
src/bin/pg_ctl/pg_ctl.c | 8 +++++-
src/gtm/gtm_ctl/gtm_ctl.c | 40 ++++++++++++++++++------------------
src/gtm/test/promote.sh | 2 +-
src/gtm/test/start_a.sh | 4 +-
src/gtm/test/start_s.sh | 4 +-
src/gtm/test/stop.sh | 4 +-
src/gtm/test2/status_a.sh | 2 +-
src/gtm/test2/status_s.sh | 2 +-
src/test/regress/pg_regress.c | 2 +-
10 files changed, 41 insertions(+), 37 deletions(-)
hooks/post-receive
--
Postgres-XC
|
|
From: Michael P. <mic...@us...> - 2012-06-03 11:59:01
|
Project "Postgres-XC".
The branch, REL1_0_STABLE has been updated
via dc3d84756ca6d38dedc48e24813a9adcd9a6a241 (commit)
via 873d1c11ea25f24815861c2e474fc43500d561c8 (commit)
via 8e61ded616bdce65c43254b010f801f39b29ce3b (commit)
via 7651ec79879d36a68c9ab32a700ee034f7367701 (commit)
via 8620f6f18eee348075f7e3430c04d7784edf1b5c (commit)
via 3dfaea4727fb3d98e28f4cb9d707c33e2bdbfa7d (commit)
via 56901beb806bde7af419ad67f49086aaca4fc8df (commit)
via d72a3bd75813615898db27d4bbd2dca5cac9615a (commit)
via fea3ea1f23fb7f72aad28a55a09a60897a08dd3d (commit)
via 8db2e820c96a0ce33d3c15f476aa78bebce2c732 (commit)
via cc2469b0925812424efc8aa24234fc3090a5e85e (commit)
via 2791f2180d8e7c469fac2a4d88fc449689e4c970 (commit)
via 6c1bf45ea50a652d2b8efa778cabb63fdf422e70 (commit)
via 71e985a07a7ae9319c00629f319a4c30a9e32fd6 (commit)
via 1a93588ffcad3e4876c9610c0a35d79c7fcdeffe (commit)
via 4615d2ca28fe9dff6e5ebaebd6caf190a0c78ba5 (commit)
via 276bea211f08a3b0d6d5753b654322c8efb781d1 (commit)
via ae7c8b290f525504978c01d77366a753505d10d5 (commit)
via 9255d21d2fe5209c96b712f96c799fe594f00f4b (commit)
via 435a1437961d72c98a1dbd27c8a9fb75024730f9 (commit)
via 69c058ffc8949c6d86240b096196c4b6b64d87e0 (commit)
via 436af241c88a2f9b971dbe248e6ec6360438033c (commit)
via 2c1382d96c575a79a2ce014dda90fd1de78b3612 (commit)
via 2ca6bac892aa650154f591c730359a968bea726a (commit)
via f93e45d4071af89cf1b37c76fbd5976d11d77347 (commit)
via 1a4bc2db6cb80a6e52aed8f5a68b895231653d6c (commit)
via 153b1dbaa0b94143041ab1e0cf249a02380f0acf (commit)
via 1b48f6af315e9db89aaf43e8c8c9de903294fbdf (commit)
via 71ca53608de1763c0d82bf463580a30f634eba56 (commit)
via 05a86891232e1bd5e43d559f30bec2fc00131f37 (commit)
via e7c43ddc266172f409f5d5bb3e571cafe7d14fa2 (commit)
from 64c7d4d0aa32515db89069827e75a4997122b4bf (commit)
- Log -----------------------------------------------------------------
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=dc3d84756ca6d38dedc48e24813a9adcd9a6a241
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=873d1c11ea25f24815861c2e474fc43500d561c8
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=8e61ded616bdce65c43254b010f801f39b29ce3b
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=7651ec79879d36a68c9ab32a700ee034f7367701
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=8620f6f18eee348075f7e3430c04d7784edf1b5c
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=3dfaea4727fb3d98e28f4cb9d707c33e2bdbfa7d
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=56901beb806bde7af419ad67f49086aaca4fc8df
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=d72a3bd75813615898db27d4bbd2dca5cac9615a
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=fea3ea1f23fb7f72aad28a55a09a60897a08dd3d
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=8db2e820c96a0ce33d3c15f476aa78bebce2c732
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=cc2469b0925812424efc8aa24234fc3090a5e85e
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=2791f2180d8e7c469fac2a4d88fc449689e4c970
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=6c1bf45ea50a652d2b8efa778cabb63fdf422e70
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=71e985a07a7ae9319c00629f319a4c30a9e32fd6
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=1a93588ffcad3e4876c9610c0a35d79c7fcdeffe
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=4615d2ca28fe9dff6e5ebaebd6caf190a0c78ba5
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=276bea211f08a3b0d6d5753b654322c8efb781d1
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=ae7c8b290f525504978c01d77366a753505d10d5
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=9255d21d2fe5209c96b712f96c799fe594f00f4b
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=435a1437961d72c98a1dbd27c8a9fb75024730f9
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=69c058ffc8949c6d86240b096196c4b6b64d87e0
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=436af241c88a2f9b971dbe248e6ec6360438033c
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=2c1382d96c575a79a2ce014dda90fd1de78b3612
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=2ca6bac892aa650154f591c730359a968bea726a
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=f93e45d4071af89cf1b37c76fbd5976d11d77347
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=1a4bc2db6cb80a6e52aed8f5a68b895231653d6c
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=153b1dbaa0b94143041ab1e0cf249a02380f0acf
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=1b48f6af315e9db89aaf43e8c8c9de903294fbdf
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=71ca53608de1763c0d82bf463580a30f634eba56
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=05a86891232e1bd5e43d559f30bec2fc00131f37
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=e7c43ddc266172f409f5d5bb3e571cafe7d14fa2
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
-----------------------------------------------------------------------
Summary of changes:
configure | 3 +-
contrib/citext/citext--unpackaged--1.0.sql | 97 ++-
contrib/pg_standby/pg_standby.c | 4 +-
contrib/pg_stat_statements/pg_stat_statements.c | 22 +-
contrib/pgcrypto/crypt-des.c | 3 +-
contrib/sepgsql/expected/label.out | 2 +-
doc/bug.template | 2 +-
doc/src/sgml/release-8.3.sgml | 229 ++++
doc/src/sgml/release-8.4.sgml | 285 ++++
doc/src/sgml/release-9.0.sgml | 320 +++++
doc/src/sgml/release-9.1.sgml | 506 +++++++-
src/backend/access/gist/gist.c | 123 ++-
src/backend/access/heap/heapam.c | 7 +
src/backend/access/index/genam.c | 11 +-
src/backend/access/nbtree/README | 8 +-
src/backend/access/nbtree/nbtpage.c | 32 +-
src/backend/access/nbtree/nbtxlog.c | 6 +-
src/backend/access/transam/xact.c | 19 +-
src/backend/catalog/heap.c | 32 +-
src/backend/catalog/objectaddress.c | 4 +
src/backend/commands/tablecmds.c | 4 +
src/backend/commands/tablespace.c | 15 +-
src/backend/po/de.po | 1415 ++++++++++----------
src/backend/po/es.po | 1378 ++++++++++----------
src/backend/po/fr.po | 1644 ++++++++++++-----------
src/backend/po/ru.po | 2 -
src/backend/po/zh_CN.po | 2 -
src/backend/po/zh_TW.po | 1 -
src/backend/port/win32_sema.c | 20 +-
src/backend/postmaster/postmaster.c | 12 +-
src/backend/tsearch/to_tsany.c | 7 +-
src/backend/utils/adt/date.c | 5 +-
src/backend/utils/adt/datetime.c | 6 +-
src/backend/utils/adt/name.c | 10 +-
src/backend/utils/adt/varchar.c | 14 +-
src/backend/utils/adt/varlena.c | 12 +-
src/backend/utils/fmgr/fmgr.c | 15 +-
src/bin/initdb/po/de.po | 6 +-
src/bin/initdb/po/es.po | 17 -
src/bin/initdb/po/fr.po | 2 -
src/bin/initdb/po/ru.po | 2 -
src/bin/initdb/po/zh_CN.po | 4 +-
src/bin/pg_basebackup/po/cs.po | 1 -
src/bin/pg_basebackup/po/de.po | 82 +-
src/bin/pg_basebackup/po/es.po | 80 +-
src/bin/pg_basebackup/po/fr.po | 92 +-
src/bin/pg_config/po/de.po | 2 -
src/bin/pg_config/po/es.po | 2 -
src/bin/pg_config/po/fr.po | 2 -
src/bin/pg_config/po/ru.po | 2 -
src/bin/pg_controldata/po/cs.po | 2 -
src/bin/pg_controldata/po/de.po | 2 -
src/bin/pg_controldata/po/es.po | 2 -
src/bin/pg_controldata/po/fr.po | 2 -
src/bin/pg_controldata/po/ru.po | 2 -
src/bin/pg_ctl/po/de.po | 2 -
src/bin/pg_ctl/po/es.po | 13 -
src/bin/pg_ctl/po/fr.po | 2 -
src/bin/pg_ctl/po/ru.po | 2 -
src/bin/pg_dump/common.c | 18 +-
src/bin/pg_dump/pg_dump.c | 39 +-
src/bin/pg_dump/pg_dump.h | 1 +
src/bin/pg_dump/po/cs.po | 2 -
src/bin/pg_dump/po/de.po | 11 +-
src/bin/pg_dump/po/es.po | 2 -
src/bin/pg_dump/po/fr.po | 4 +-
src/bin/pg_dump/po/ru.po | 2 -
src/bin/pg_dump/po/sv.po | 2 -
src/bin/pg_resetxlog/po/cs.po | 2 -
src/bin/pg_resetxlog/po/de.po | 2 -
src/bin/pg_resetxlog/po/es.po | 2 -
src/bin/pg_resetxlog/po/fr.po | 2 -
src/bin/pg_resetxlog/po/ru.po | 2 -
src/bin/psql/describe.c | 7 +
src/bin/psql/po/de.po | 169 ++--
src/bin/psql/po/es.po | 157 ++-
src/bin/psql/po/fr.po | 207 ++--
src/bin/psql/po/ru.po | 2 -
src/bin/psql/po/zh_TW.po | 1 -
src/bin/scripts/po/cs.po | 2 -
src/bin/scripts/po/de.po | 19 +-
src/bin/scripts/po/es.po | 8 -
src/bin/scripts/po/fr.po | 2 -
src/bin/scripts/po/ru.po | 2 -
src/include/catalog/heap.h | 2 +
src/include/pg_config.h.win32 | 6 +-
src/include/utils/timestamp.h | 10 +
src/interfaces/ecpg/ecpglib/po/de.po | 2 -
src/interfaces/ecpg/ecpglib/po/es.po | 2 -
src/interfaces/ecpg/ecpglib/po/fr.po | 2 -
src/interfaces/ecpg/preproc/po/de.po | 2 -
src/interfaces/ecpg/preproc/po/es.po | 14 -
src/interfaces/ecpg/preproc/po/fr.po | 2 -
src/interfaces/libpq/libpq.rc.in | 8 +-
src/interfaces/libpq/po/cs.po | 2 -
src/interfaces/libpq/po/de.po | 133 +-
src/interfaces/libpq/po/es.po | 119 +-
src/interfaces/libpq/po/fr.po | 167 ++--
src/interfaces/libpq/po/it.po | 1 -
src/interfaces/libpq/po/ru.po | 2 -
src/interfaces/libpq/po/sv.po | 2 -
src/interfaces/libpq/po/zh_CN.po | 2 -
src/interfaces/libpq/po/zh_TW.po | 1 -
src/pl/plperl/po/cs.po | 1 -
src/pl/plperl/po/de.po | 2 -
src/pl/plperl/po/es.po | 7 -
src/pl/plperl/po/fr.po | 2 -
src/pl/plpgsql/src/pl_exec.c | 1 +
src/pl/plpgsql/src/po/es.po | 2 -
src/pl/plpgsql/src/po/fr.po | 2 -
src/pl/plpython/po/de.po | 167 ++-
src/pl/plpython/po/es.po | 190 ++--
src/pl/plpython/po/fr.po | 176 ++--
src/pl/tcl/po/de.po | 2 -
src/pl/tcl/po/es.po | 2 -
src/pl/tcl/po/fr.po | 2 -
src/port/win32ver.rc | 4 +-
src/test/regress/input/security_label.source | 1 +
src/test/regress/output/security_label.source | 2 +
src/timezone/data/africa | 36 +-
src/timezone/data/antarctica | 21 +-
src/timezone/data/asia | 72 +-
src/timezone/data/australasia | 28 +-
src/timezone/data/europe | 14 +-
src/timezone/data/leapseconds | 42 +-
src/timezone/data/northamerica | 135 ++-
src/timezone/data/southamerica | 53 +-
src/timezone/data/zone.tab | 5 +-
128 files changed, 5253 insertions(+), 3462 deletions(-)
hooks/post-receive
--
Postgres-XC
|
|
From: Michael P. <mic...@us...> - 2012-06-02 20:38:51
|
Project "Postgres-XC".
The branch, master has been updated
via dfee5379109a757e1bfaa9fe59830e6fb05f6688 (commit)
from af488de7241de630c68c2f8a7a5ec6efb68ad083 (commit)
- Log -----------------------------------------------------------------
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=dfee5379109a757e1bfaa9fe59830e6fb05f6688
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
-----------------------------------------------------------------------
Summary of changes:
src/backend/postmaster/postmaster.c | 77 +++++++++++++++++++++++------------
1 files changed, 51 insertions(+), 26 deletions(-)
hooks/post-receive
--
Postgres-XC
|
|
From: Michael P. <mic...@us...> - 2012-06-02 20:38:47
|
Project "Postgres-XC".
The branch, REL1_0_STABLE has been updated
via 64c7d4d0aa32515db89069827e75a4997122b4bf (commit)
from 93aae0131c08cc5204562e7ce335f2446fc431af (commit)
- Log -----------------------------------------------------------------
http://postgres-xc.git.sourceforge.net/git/gitweb.cgi?p=postgres-xc/postgres-xc;a=commitdiff;h=64c7d4d0aa32515db89069827e75a4997122b4bf
commit dfee5379109a757e1bfaa9fe59830e6fb05f6688
Author: Michael Paquier <mi...@ot...>
Date: Sun Jun 3 05:29:06 2012 +0900
Block node registration for slave nodes and perform register at a stable state
Performing node registration on GTM just after postmaster start is dangerous
because the master/slave status of the node is unknown. So node
registration is done once node has reached a stable state, meaning that
Xlog process has confirmed to postmaster that recovery has been completed.
Registration is also blocked for slave nodes. A standby node is created
from a hot backup of its master so master and its slaves will share the
same node name. Two nodes are not authorized to register to GTM with the
same node name to avoid identification confusion in the cluster, so it
looks normal to let the master and the master only identify itself.
A a direct consequence, unregistration too is done only for master node.
Synchronous and asynchronous node replication works correctly with this fix,
as well as node promotion (Ex: "pg_ctl promote -D $DATA").
M src/backend/postmaster/postmaster.c
-----------------------------------------------------------------------
Summary of changes:
src/backend/postmaster/postmaster.c | 77 +++++++++++++++++++++++------------
1 files changed, 51 insertions(+), 26 deletions(-)
hooks/post-receive
--
Postgres-XC
|
