eb0af7ca23
Always pass -D/--pgdata option to pg_basebackup
...
repmgr does not require explicit provision of the target data
directory when cloning a standby (it defaults to the same directory
as on the master). However this is a required option for pg_basebackup
which was only being provided if repmgr's -D/--data-dir option was
set, so ensure we always provide whatever repmgr is using.
Per report from Martín.
2015-08-25 14:36:51 +09:00
ae47e5f413
Fix example command syntax
2015-08-12 12:32:41 +09:00
46100a9549
"Supports 9.3 and 9.4" -> "Supports 9.3 and later"
...
In case anyone thinks it might not run on 9.5.
2015-08-12 12:26:27 +09:00
9bd95cabdf
Merge pull request #98 from gciolli/master
...
Bug #97 fix (standby clone --force does not empty pg_xlog).
2015-08-11 15:10:37 +02:00
f1584469bf
Bug #97 fix (standby clone --force does not empty pg_xlog).
2015-08-11 14:59:23 +02:00
a7f46d24de
Merge pull request #96 from gciolli/master
...
Bug #90 fix (autofailover with reconnect_attemps > 1).
2015-08-11 06:35:31 +05:30
462d446477
Bug #90 fix (autofailover with reconnect_attemps > 1).
...
The main change is that now check_connection requires a conninfo
parameter, and the connection object has type (PGconn **) so it can be
replaced by check_connection if needed.
The bug was caused by the fact that the first failure resulted in
*conn == NULL, so that subsequent checks of the upstream connection
were failing irrespectively of the actual state of the upstream node.
Now, when *conn == NULL, check_connection will use conninfo to
establish a new connection and place it into *conn. We introduce a new
INTERNAL_ERROR code for the case when they are both NULL.
In passing, we also reworded a confusing error message, distinguishing
a timeout from the actual elapsed time.
2015-08-10 20:58:43 +02:00
23a72f489c
Add note about configuring 'shared_preload_libraries' for repmgr
...
Useful to put all the possible postgresql.conf changes in one place.
2015-08-07 14:37:57 +09:00
f3f56b0cd6
fix repmgrd spelling in default config files
...
make sure the daemon has it's trailing _d_ everywhere
2015-08-06 17:20:56 +02:00
00146b7fbd
Merge pull request #88 from soxwellfb/fix-config-reader
...
Ignore comments after values
2015-08-04 13:28:04 +05:30
faf72a2514
Ignore comments after values
2015-07-15 14:53:35 +10:00
7010b636e0
Reword error message for consistency
...
Github issue #87
2015-07-10 10:11:23 +09:00
00deff9069
[Fix] if 'node' config parameter is set as a string, nothing complain and all nodes are set to id '0'.
2015-07-06 11:20:31 +02:00
5240a5723a
Clarify WAL file retention issue during standby cloning.
...
Per GitHub issue #86 .
2015-07-02 09:32:30 +09:00
899d789699
Fixed version number and package dependencies in deb control file.
2015-06-04 11:26:36 +02:00
cd7a3215df
Fixed tabs/spaces issues caused by lazy copy/paster in github web UI when creating the PR.
2015-06-04 11:25:45 +02:00
45e29c5b28
Typo in a comment
2015-06-04 18:12:35 +09:00
f8fd344d9f
Updated makefile for deb creation
...
Added the ability to fetch the installed postgresql version when building the deb package so that install paths are correct.
2015-06-04 09:51:03 +02:00
5def293ed6
Configuration file: ignore whitespace following the '=' sign
2015-06-04 11:58:34 +09:00
ff7b4d3f02
Improve whitespace handling
...
Ignore blank lines which consist of whitespace. Per issue #71 in
GitHub.
This fix also improves comment handling and will treat lines with
whitespace before the '#' character as whitespace.
2015-06-04 11:44:38 +09:00
a54478a045
Documentation tweaks
2015-05-12 10:16:52 +09:00
7ad9a2c28a
Allow repmgr to be compiled with a libpq newer than PostgreSQL
...
Fixes #44
2015-04-29 18:00:54 +02:00
3deb6784e7
Fixed typo
2015-04-16 13:30:40 +09:00
ba275bb0c2
Update HISTORY file
2015-04-16 11:02:40 +09:00
9735bb63a1
Add short option '-c' for '--fast-checkpoint'
...
This is similar to pg_basebackup's '-c/--checkpoint=fast|spread'
option.
2015-04-16 08:15:44 +09:00
1e5792f8df
Remove unused function
2015-04-14 14:29:47 +09:00
a01fefa7d0
After standby promotion, ensure metadata is updated by repmgr
...
Previously this was handled by repmgrd but if a standby is promoted
directly this will leave the metadata in an incorrect state.
2015-04-14 13:39:48 +09:00
34eaf94b2b
Fix variable name spelling
2015-04-14 13:15:44 +09:00
68e3a9d7ab
Improve promotion failure handling
...
Exit with error; don't attempt to write event log as master
connection will not be valid at this point.
2015-04-14 13:09:02 +09:00
2ad4f68700
Add notes about cluster name and quoting
...
Supported, but better to avoid.
2015-04-13 15:40:29 +09:00
00aa0c8c87
Fix typos
2015-04-13 15:22:09 +09:00
e8025c7c9f
Re-use replication slot if it already exists
...
Per issue #65 in GitHub
2015-04-13 13:17:38 +09:00
6a17360b4c
Add FAQ item about shared_preload_libraries = 'repmgr_funcs'
2015-04-10 16:45:04 +09:00
9e5e843a4f
Prevent a test SSH connection being made when not needed
...
When using the default pg_basebackup method to clone a standby
together with the `--ignore-external-config-files` option, there's
no need to test for a working SSH connection - which may not
be desirable in many use-cases anyway.
Per issue #64 in GitHub.
2015-04-10 15:05:19 +09:00
734ae1825e
Don't run ssh connection test when --ignore-external-config-files was used
2015-04-10 11:02:19 +09:00
41fe58764e
Add note about PGPASSFILE
2015-04-10 10:29:52 +09:00
58a5249b7e
HISTORY update
2015-04-09 16:12:11 +09:00
90c0bd4638
Add FAQ item about witness server port
2015-04-09 16:05:44 +09:00
359e81a6d6
Updated RHEL files.
2015-04-02 05:54:08 -03:00
0037e66034
fix messy formatting
2015-03-31 22:58:50 +05:30
07d220cb00
Correct monitoring table column names
...
It would be more consistent to change the "primary" to "master"
but that would make the table incompatible with the v2.0 table.
2015-03-31 18:14:32 +09:00
4dfeffe087
Add constant NODE_NOT_FOUND
...
Which is what the magic number means in those contexts.
2015-03-31 14:35:16 +09:00
18544c82ca
Prevent rempgrd from looping infinitely if node was not registered
2015-03-31 14:25:08 +09:00
0f86bdcd05
Fixes for event logging
...
We can't always assume a valid connection to the master
2015-03-31 14:15:29 +09:00
7d33c1e411
Only attempt to log an event if the rempgr schema has been set
...
In some circumstances (primarily when executing `repmgr standby
clone`) the `repmgr.conf` file is not mandated. However this means
the repmgr schema is not known, and any attempt to create an
event record will result in a log warning, which may cause
confusion as to the success of the operation.
It might be better to mandate providing `repmgr.conf` in all
circumstances.
Per report in https://github.com/2ndQuadrant/repmgr/issues/53 .
2015-03-31 10:28:34 +09:00
fec65bde3d
Fix typo in -?/--help output
2015-03-27 18:09:41 +09:00
4863ea98bc
Bump specfile version number
2015-03-27 17:42:55 +09:00
c4505248b0
doc: Minor README/FAQ rewording
2015-03-27 11:34:52 +09:00
5774d86ac7
Remove not-yet-existent 2.0.3 release items
2015-03-27 11:05:40 +09:00
29e7733152
Prepare devel branch for promotion to master
2015-03-26 11:50:22 +09:00
Ian Barwick