e40b9db0a6
Properly specify rsync --exclude directories
...
Using '--exclude=dirname/*' to explicitly specify directories whose contents
should not be copied. This will result in empty directories being created
on the destination if they exist on the source, but that's not a problem as
they are needed anyway.
Previously the generated rsync command contained '--exclude=pg_log*', which
will break replication on 9.5 as the wildcard expansion prevents the
'pg_logical' directory from being copied.
2014-07-07 10:42:45 +09:00
54e62c3d65
Typo fixes and minor wording tweaks for clarity
2014-07-03 15:41:51 +09:00
bfd482bebc
Merge pull request #28 from Nexperteam/rhel-scripts
...
adapt makefile for RHEL + RHEL specific files
2014-06-26 22:55:56 +05:30
6a0fc43086
Merge pull request #26 from Nexperteam/master
...
Makefile: create bindir before instal + force dir
2014-06-26 22:54:09 +05:30
8f47111072
Merge pull request #31 from riegie/patch-1
...
Update README.rst
2014-06-26 22:51:52 +05:30
0b5b3aaa4b
Update README.rst
...
Fixing a grammar mistake.
2014-06-19 20:53:42 -04:00
d8bba0de03
init script: make status call return proper return code
2014-06-06 15:47:33 +02:00
a11bb60fd5
re-add comment full debug of log.c
2014-06-06 14:45:27 +02:00
4f810aee2c
on init of witness server create db and user to avoid using postgres
2014-06-06 14:36:12 +02:00
73d352b2a2
adapt makefile for RHEL + RHEL specific files
2014-06-06 14:30:37 +02:00
143aa57bb8
Makefile: create bindir before instal + force dir
...
in order to facilitate building of an rpm I've added an / to the end of the dirs.
2014-04-28 14:27:52 +02:00
5b15fcff5c
Merge pull request #24 from PriceChild/patch-1
...
Typo in example command.
2014-04-08 11:04:09 +05:30
4469de533e
Typo in example command.
2014-04-01 16:31:33 +01:00
30045f3bec
keep naming consistent
2014-03-24 15:18:37 +00:00
d93d42fadb
fix: witness creation and monitoring
...
While reading node entries from master use a separate PGresult when inserting into witness.
Witness monitoring supplies a null value for 'last_apply_time'.
2014-03-24 14:19:29 +00:00
7c89a4d762
Merge pull request #20 from kjoe/master
...
debian init script and config file documentation fixes
2014-03-24 07:46:56 +01:00
b3c68dead8
debian init script and config file documentation fixes
2014-03-22 22:23:15 +01:00
b9ab9010c0
debian init script and config file documentation fixes
2014-03-22 22:13:33 +01:00
2a6c835a5a
debian init script and config file documentation fixes
2014-03-21 14:18:07 +01:00
2d48d5aee4
fixing some documentation errors
2014-03-10 15:48:44 +01:00
653e11c2a7
basic min_recovery_apply_delay support
2014-03-10 15:41:38 +01:00
91c29fe2a2
removed old comment
2014-03-06 18:34:41 +01:00
573f1d3b2e
no longer use global variable for SQL query buffer
2014-03-06 18:34:41 +01:00
0a6ff7faec
removed no-longer used variable
2014-03-06 18:34:41 +01:00
98b1f8d28a
rather big refactoring: use a naming scheme
...
In the past naming of functions, variables and such didn't really have a
naming scheme. Now they should have.
2014-03-06 18:34:40 +01:00
9eba986833
avoid usage of snprintf()
...
We have a nice little abstraction for snprintf with covering the case
that a string is too big for the target buffer – let's use that!
2014-03-06 18:34:40 +01:00
164cf9d08f
completely avoid usage of strnlen()
2014-03-06 18:34:40 +01:00
d8b8bf0e2a
pg_indent'ing all files…
2014-03-06 18:34:40 +01:00
fed5c77653
various improvements and bugfixes in the init script
2014-03-06 15:23:22 +01:00
8429b43edf
Merge pull request #14 from wamonite/fix_follow_user
...
fix: store the master connection user name on standby follow
2014-03-06 15:20:02 +01:00
7e55ce737d
fix: store the master connection user name on standby follow
2014-03-05 16:49:56 +00:00
98c7635fb5
fixing more compiler warnings
2014-03-04 17:58:36 +01:00
90ecb2b107
fix: check return values of freopen()
...
Some compiles complain about not checking the return value of freopen(),
so we check it
2014-03-04 15:32:48 +01:00
50b9022a41
fix: don't use Windows newlines
2014-03-04 12:59:23 +01:00
150ccc0662
add option to avoid repmgrd started upon installation
...
Now repmgr.repmgrd.default has another option: REPMGRD_ENABLED. Valid
values are either yes or no.
2014-03-04 12:46:05 +01:00
0ff14a2aa1
avoid compiler warnings
2014-02-21 13:47:29 +01:00
5215265694
fix: now CloseConnections() is much more safe
2014-02-18 17:06:36 +01:00
e45ac25348
fix: progname is const, do not free it
...
The leak is irrelevant
2014-02-18 16:45:35 +01:00
a1ce01f033
fix: fixed some leaks
2014-02-18 16:35:29 +01:00
516cde621a
fix: strcpy() on overlapping memory regions is invalid
2014-02-18 15:42:20 +01:00
f0807923a3
fix: gettimeofday() expects two arguments
2014-02-18 15:33:56 +01:00
10ca8037f8
added some more log messages
...
Now we should be able to distinguish different events more easily
2014-02-18 14:10:12 +01:00
0dc46f0dc8
fix: set connection to NULL when finishing it
...
This will avoid CloseConnections() to try to close an already closed connection.
2014-02-18 13:42:49 +01:00
c3b58658ad
fixing repmgr repl_status columns
...
repmgr repl_status had the column time_lag which was documented to be
the time a standby is behind master. In fact it only works like this
when viewed on the standby and not on the master: there it only was the
time of the last status update. We dropped that column and replaced it
by a new column „communication_time_lag“ which is the content of the
repl_status column on the master. On the standby we contain the time of
the last update in shared mem though refer always to the correct time
nonetheless where repl_status is queried. We also added a new column,
„replication_time_lag“, which refers to the apply delay.
2014-02-15 01:35:27 +01:00
18f1fed77f
fixing wait_connection_availability()
...
wait_connection_availability() did take at least 2 seconds per call in
the old incarnation. Now we may finish a call without any sleep at all
when the result is already ready at the time called
2014-02-15 01:31:12 +01:00
d58fd080ca
flush stderr after a log message appears
...
We had the problem that the log file appeared empty for a long time due
to file buffers. Thus we call fflush() after every log message so the
log file gets written out to disk quickly
2014-02-15 01:29:12 +01:00
c4ac2d3343
fixing PQexec() calls
...
fixing several calls where we did not check the result status but only
the return value; the query may fail nonetheless
2014-02-15 01:27:53 +01:00
5ff1beeea7
do not enable autofailover by default
...
Autofailover is an experimental feature which should not be enabled by
default. The user has to be aware of what he is doing when enabling it.
2014-02-11 09:27:31 +01:00
ca470647cb
cleanup of usage text
...
Now it properly aligns and breaks at 78 characters.
2014-01-30 14:26:17 +01:00
62ee287e3f
updated TODO
2014-01-30 14:10:14 +01:00
Ian Barwick