1
$Id: RELEASE-1.2.2,v 1.1.2.4 2007-01-05 20:50:08 cbbrowne Exp $
3
Release 1.2.2 of Slony-I has the following bug fixes over 1.2.1:
5
- SGML tagging fixes to allow docs to be generated on Fedora
7
- Fixes to altperl scripts (init cluster, store node) so they would
8
properly generate STORE PATH requests (which had broken when STORE
9
LISTEN code was removed)
11
- If PostgreSQL version is 8.1 or greater, we can use pg_config
12
sharedir. Otherwise we will resort to the default guessing mode.
14
- Some makefile hygenic fixes
16
- Added \n to a number of log requests
18
- Fix to UPDATE FUNCTIONS which did not consider versions 1.1.5/1.1.6
19
to be members of the 1.1 stream. The minor problem with this was
20
that Slony-I tables wouldn't get altered to drop out OIDs. The
21
major problem was that the new table sl_registry wouldn't get
24
- Removed some confusing NOTICEs concerning management of partial
25
indexes on sl_log_1 and sl_log_2.
27
- Close file descriptors upon running slon_terminate_worker() so that
28
we don't run out of file descriptors
30
- Added in sleep(10) requests in several places where attempts to
31
access the database in the main thread fails; this means that, for
32
instance, the main DB is down, the slon doesn't simply sit there
33
continually attempting to reconnect, as many times per second as it
36
- Several fixes to RPM .spec files
38
- Environment fix to Win32 service
40
- Set escape_string_warning to off, to prevent pgsql log noise when
41
replicating cross versions
43
- Fix memory leak: free logshipping query
45
- Bug #1585 - on 8.0+, if TRUNCATE failed, indices were not being
46
deactivated during SUBSCRIBE SET COPY request, greatly slowing loading
47
of data. Added a deactivation inside the exception block, so that
48
indices are deactivated during the COPY no matter what.
50
- X-Fade noticed that UPGRADE FUNCTIONS was taking out exclusive locks
51
on sl_log_1, sl_log_2, sl_seqlog, which means that doing a Slony-I
52
upgrade involved imposing an application outage.
54
Moved the ALTER TABLE ... WITHOUT OIDS requests for these tables to
55
cleanup loops, elsewhere, so that these tables no longer need to be
56
locked as part of the upgrade process.
58
- Documentation changes: Need for client/server encodings to match
60
- Fixes to parameter handling and interpreter name for
61
test_slony_state*.pl scripts
63
- There was an error condition where if the final line of a slon.conf
64
file did not end with a \n, then:
65
a) A comment on that line would be treated as a syntax error
66
b) A value set on that line would be ignored