c918aaad4a
Update TODO
2015-10-23 08:49:44 +09:00
6e7eee4c01
Only log some debug items if verbose flag is set.
2015-10-23 08:29:35 +09:00
5c59e8fc5b
Add missing space
2015-10-22 13:13:01 +09:00
eba0b6bb1e
Update TODO
2015-10-14 16:57:24 +09:00
3bc0b80a71
Reword log level error message to be more like the Postgres one
2015-10-07 10:54:47 +09:00
06b9e0a8ec
Clarify purpose of get_repmgr_schema()
2015-10-07 10:54:47 +09:00
120be2db1c
Fix bug which prevents repmgrd from starting when the cluster name has
...
upper case letters.
2015-10-06 20:54:28 -03:00
12bd7da836
Clean up --help output
...
There are a confusing number of command line options, some
of which are only valid for particular operations, e.g. "standby clone".
2015-10-05 11:53:45 +09:00
2fd905cf9e
Update HISTORY
2015-10-02 14:40:52 +09:00
7a8a50e229
Add CONTRIBUTING.md
2015-10-02 10:18:06 +09:00
e188044593
Clean up markup
2015-10-02 10:07:36 +09:00
636f4b03c6
Spelling and markup fixes
2015-10-02 09:55:34 +09:00
bf96b383a3
Merge branch 'eternaltyro-master'
2015-10-02 09:49:55 +09:00
3a2e40f381
Merge branch 'master' of https://github.com/eternaltyro/repmgr into eternaltyro-master
2015-10-02 09:49:39 +09:00
c608bb28ee
Fix some typos
2015-10-02 08:03:55 +09:00
ca9c2e1143
Fix -D/--data-dir's long option
...
This was erroneously rendered as --dest-dir. This bugfeature seems
to have been around for a long time; however as the only way anyone
could know of the existence of --dest-dir is by reading the source
code, we can safely remove it.
Pointed out by Github user Jehan-Guillaume (ioguix) de Rorthais.
2015-10-02 07:50:19 +09:00
3a6d6b8899
Fix typo in README.md
...
Reported by Github user knopwob.
2015-10-02 07:17:39 +09:00
4091cb7f18
Update TODO
2015-10-01 16:09:37 +09:00
870b0a53b6
Allow 'primary' as synonym for 'master'.
...
"Primary" is the term preferred in the PostgreSQL documentation, so
we should at least support it.
Practically this means it's possible to write "rempgr primary register"
in place of "repmgr master register".
The next feature-release should replace "master" with "primary" in
the documentation and log messages.
Per gripe in Github #112 .
2015-10-01 16:01:12 +09:00
6184cc57be
Bump version in HEAD
2015-10-01 13:37:23 +09:00
e1254b6773
Update HISTORY
2015-10-01 13:24:02 +09:00
1c9121c2d8
Witness server - extract explicit port setting from conninfo setting
...
This makes the '-l/--local-port' option redundant, which is now
marked as deprecated.
2015-10-01 09:42:44 +09:00
6da03a6157
Free connection options array after use.
2015-10-01 09:16:36 +09:00
9bb6befa25
Sanity check 'conninfo' parameter
...
Catch errors early, it makes everyone's life easier.
2015-10-01 09:06:27 +09:00
a8e5c68d03
Better document '-l/--local-port' option.
...
Per Github #59 .
2015-10-01 08:40:27 +09:00
b83e18c503
Fix '-l, --local-port' help output.
...
Is for witness server only.
Use constants for default port definitions while we're at it.
Github #123 .
2015-10-01 08:24:14 +09:00
d4b845d213
'repmgr witness create --force': overwrite any existing node record
...
Consistent with 'repmgr standby register --force'.
Fixes Github #122 .
2015-10-01 08:02:06 +09:00
75aad9a85e
repmgr witness create --force: overwrite existing data directory
...
Per Github #82 (dimitri).
2015-09-30 17:05:54 +09:00
e115825cd6
Fix comment capitalization
2015-09-30 14:58:43 +09:00
6cf5ab2e53
Update HISTORY
2015-09-30 12:48:10 +09:00
f8119d20ea
Handle tablespace mapping in 9.5 and later in --rsync-only mode
...
9.5 introduces the tablespace_map file, which is created on the upstream
node while a backup is running. We need to overwrite this with the
provided values.
Note that we only write explicitly-provided values to the tablespace_map
file, however the existing symlinks for non-specified tablespaces
will have been copied anyway.
Fixes Github #119 .
2015-09-30 11:48:09 +09:00
0caddf2d2c
Fix check when tablespace mapping option used with 9.3
...
As pointed out by EvilElk (Github issue #91 ).
2015-09-29 14:30:37 +09:00
a4abbc6f0c
Minor formatting fixes
2015-09-29 14:18:48 +09:00
d7e489ea0a
Update repmgr --help output
...
(per Github report #113 from renard)
2015-09-25 11:26:12 +09:00
2bcacff3b3
Update TODO
2015-09-25 11:19:21 +09:00
45eb0ea5d3
Miscelleanous comment fixes
2015-09-25 11:17:26 +09:00
c3bd02b83d
Standardize if-statement formatting
...
"if(" -> "if ("
2015-09-24 17:45:08 +09:00
8e7d110a22
Check for existing master record before deleting it
...
Otherwise repmgr implies it's deleting a record which isn't actually
there.
2015-09-24 17:39:39 +09:00
43874d5576
Close database connection as soon as it is no longer required
2015-09-24 16:55:59 +09:00
87ff9d09ba
Ensure --force option works with "master register"
...
This was failing previously if repmgr was able to connect to the
master database
2015-09-24 16:55:59 +09:00
c429b0b186
Don't fail with error when registering master if schema already defined
...
Registering a master creates the schema, but it may be desirable
to forcibly reregister a master without deleting the schema, so
uncouple the dependency.
Also ensure schema creation is atomic by wrapping it in a transaction.
Per GitHub issue #49 .
2015-09-24 16:55:43 +09:00
dd7ebdc1c7
Add more information in "cluster show"
...
- Add display of current node name.
- Add display of upstream node name.
2015-09-23 16:02:10 +02:00
03b88178c1
Minor change in an error log message
2015-09-23 10:44:07 -03:00
5f33f4286f
Merge pull request #115 from xocolatl/unregister
...
Add a STANDBY UNREGISTER command.
2015-09-23 10:17:22 -03:00
932f84910b
Add a STANDBY UNREGISTER command.
...
In some cases it is desirable to remove repmgr's handling of a standby, perhaps
because that standby is to be removed altogether. With no UNREGISTER command,
one had to manually delete the corresponding row from repl_nodes.
2015-09-23 12:37:42 +02:00
1ef7f1368d
Update TODO
2015-09-23 14:05:09 +09:00
640abed18f
Don't close connection to master until upstream node updated
2015-09-23 13:11:51 +09:00
ef6b24551a
call update_node_record_set_upstream() for STANDBY FOLLOW
...
repmgrd correctly updates ID of the upstream node after automatic
failover, but repmgr was not doing that for manual failvers.
This moves the existing function to dbutils and modifies it so that
it does not rely on global variables with configuration (available
just in repmgrd).
This should fix issue #67 (hopefully, haven't done much testing).
2015-09-23 12:32:47 +09:00
42847e44d2
Update HISTORY file
2015-09-23 10:29:40 +09:00
dd7cfce3d3
Add TODO note about default configuration file.
...
Per suggestion in #108 .
2015-09-22 19:26:19 +09:00
Ian Barwick