Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Skip to content

Commit b940859

Browse files
tglsfdctglsfdc
committed
Release notes for 9.4.4, 9.3.9, 9.2.13, 9.1.18, 9.0.22.
1 parent f6e9cbf commit b940859

File tree

5 files changed

+363
-5
lines changed

5 files changed

+363
-5
lines changed

‎doc/src/sgml/release-9.0.sgml

+74
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,80 @@
11
<!-- doc/src/sgml/release-9.0.sgml -->
22
<!-- See header comment in release.sgml about typical markup -->
33

4+
<sect1 id="release-9-0-22">
5+
<title>Release 9.0.22</title>
6+
7+
<note>
8+
<title>Release Date</title>
9+
<simpara>2015-06-12</simpara>
10+
</note>
11+
12+
<para>
13+
This release contains a small number of fixes from 9.0.21.
14+
For information about new features in the 9.0 major release, see
15+
<xref linkend="release-9-0">.
16+
</para>
17+
18+
<para>
19+
The <productname>PostgreSQL</> community will stop releasing updates
20+
for the 9.0.X release series in September 2015.
21+
Users are encouraged to update to a newer release branch soon.
22+
</para>
23+
24+
<sect2>
25+
<title>Migration to Version 9.0.22</title>
26+
27+
<para>
28+
A dump/restore is not required for those running 9.0.X.
29+
</para>
30+
31+
<para>
32+
However, if you are upgrading from a version earlier than 9.0.18,
33+
see <xref linkend="release-9-0-18">.
34+
</para>
35+
36+
</sect2>
37+
38+
<sect2>
39+
<title>Changes</title>
40+
41+
<itemizedlist>
42+
43+
<listitem>
44+
<para>
45+
Fix rare failure to invalidate relation cache init file (Tom Lane)
46+
</para>
47+
48+
<para>
49+
With just the wrong timing of concurrent activity, a <command>VACUUM
50+
FULL</> on a system catalog might fail to update the <quote>init file</>
51+
that's used to avoid cache-loading work for new sessions. This would
52+
result in later sessions being unable to access that catalog at all.
53+
This is a very ancient bug, but it's so hard to trigger that no
54+
reproducible case had been seen until recently.
55+
</para>
56+
</listitem>
57+
58+
<listitem>
59+
<para>
60+
Avoid deadlock between incoming sessions and <literal>CREATE/DROP
61+
DATABASE</> (Tom Lane)
62+
</para>
63+
64+
<para>
65+
A new session starting in a database that is the target of
66+
a <command>DROP DATABASE</> command, or is the template for
67+
a <command>CREATE DATABASE</> command, could cause the command to wait
68+
for five seconds and then fail, even if the new session would have
69+
exited before that.
70+
</para>
71+
</listitem>
72+
73+
</itemizedlist>
74+
75+
</sect2>
76+
</sect1>
77+
478
<sect1 id="release-9-0-21">
579
<title>Release 9.0.21</title>
680

‎doc/src/sgml/release-9.1.sgml

+68
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,74 @@
11
<!-- doc/src/sgml/release-9.1.sgml -->
22
<!-- See header comment in release.sgml about typical markup -->
33

4+
<sect1 id="release-9-1-18">
5+
<title>Release 9.1.18</title>
6+
7+
<note>
8+
<title>Release Date</title>
9+
<simpara>2015-06-12</simpara>
10+
</note>
11+
12+
<para>
13+
This release contains a small number of fixes from 9.1.17.
14+
For information about new features in the 9.1 major release, see
15+
<xref linkend="release-9-1">.
16+
</para>
17+
18+
<sect2>
19+
<title>Migration to Version 9.1.18</title>
20+
21+
<para>
22+
A dump/restore is not required for those running 9.1.X.
23+
</para>
24+
25+
<para>
26+
However, if you are upgrading from a version earlier than 9.1.16,
27+
see <xref linkend="release-9-1-16">.
28+
</para>
29+
30+
</sect2>
31+
32+
<sect2>
33+
<title>Changes</title>
34+
35+
<itemizedlist>
36+
37+
<listitem>
38+
<para>
39+
Fix rare failure to invalidate relation cache init file (Tom Lane)
40+
</para>
41+
42+
<para>
43+
With just the wrong timing of concurrent activity, a <command>VACUUM
44+
FULL</> on a system catalog might fail to update the <quote>init file</>
45+
that's used to avoid cache-loading work for new sessions. This would
46+
result in later sessions being unable to access that catalog at all.
47+
This is a very ancient bug, but it's so hard to trigger that no
48+
reproducible case had been seen until recently.
49+
</para>
50+
</listitem>
51+
52+
<listitem>
53+
<para>
54+
Avoid deadlock between incoming sessions and <literal>CREATE/DROP
55+
DATABASE</> (Tom Lane)
56+
</para>
57+
58+
<para>
59+
A new session starting in a database that is the target of
60+
a <command>DROP DATABASE</> command, or is the template for
61+
a <command>CREATE DATABASE</> command, could cause the command to wait
62+
for five seconds and then fail, even if the new session would have
63+
exited before that.
64+
</para>
65+
</listitem>
66+
67+
</itemizedlist>
68+
69+
</sect2>
70+
</sect1>
71+
472
<sect1 id="release-9-1-17">
573
<title>Release 9.1.17</title>
674

‎doc/src/sgml/release-9.2.sgml

+68
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,74 @@
11
<!-- doc/src/sgml/release-9.2.sgml -->
22
<!-- See header comment in release.sgml about typical markup -->
33

4+
<sect1 id="release-9-2-13">
5+
<title>Release 9.2.13</title>
6+
7+
<note>
8+
<title>Release Date</title>
9+
<simpara>2015-06-12</simpara>
10+
</note>
11+
12+
<para>
13+
This release contains a small number of fixes from 9.2.12.
14+
For information about new features in the 9.2 major release, see
15+
<xref linkend="release-9-2">.
16+
</para>
17+
18+
<sect2>
19+
<title>Migration to Version 9.2.13</title>
20+
21+
<para>
22+
A dump/restore is not required for those running 9.2.X.
23+
</para>
24+
25+
<para>
26+
However, if you are upgrading from a version earlier than 9.2.11,
27+
see <xref linkend="release-9-2-11">.
28+
</para>
29+
30+
</sect2>
31+
32+
<sect2>
33+
<title>Changes</title>
34+
35+
<itemizedlist>
36+
37+
<listitem>
38+
<para>
39+
Fix rare failure to invalidate relation cache init file (Tom Lane)
40+
</para>
41+
42+
<para>
43+
With just the wrong timing of concurrent activity, a <command>VACUUM
44+
FULL</> on a system catalog might fail to update the <quote>init file</>
45+
that's used to avoid cache-loading work for new sessions. This would
46+
result in later sessions being unable to access that catalog at all.
47+
This is a very ancient bug, but it's so hard to trigger that no
48+
reproducible case had been seen until recently.
49+
</para>
50+
</listitem>
51+
52+
<listitem>
53+
<para>
54+
Avoid deadlock between incoming sessions and <literal>CREATE/DROP
55+
DATABASE</> (Tom Lane)
56+
</para>
57+
58+
<para>
59+
A new session starting in a database that is the target of
60+
a <command>DROP DATABASE</> command, or is the template for
61+
a <command>CREATE DATABASE</> command, could cause the command to wait
62+
for five seconds and then fail, even if the new session would have
63+
exited before that.
64+
</para>
65+
</listitem>
66+
67+
</itemizedlist>
68+
69+
</sect2>
70+
</sect1>
71+
472
<sect1 id="release-9-2-12">
573
<title>Release 9.2.12</title>
674

‎doc/src/sgml/release-9.3.sgml

+149
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,155 @@
11
<!-- doc/src/sgml/release-9.3.sgml -->
22
<!-- See header comment in release.sgml about typical markup -->
33

4+
<sect1 id="release-9-3-9">
5+
<title>Release 9.3.9</title>
6+
7+
<note>
8+
<title>Release Date</title>
9+
<simpara>2015-06-12</simpara>
10+
</note>
11+
12+
<para>
13+
This release contains a small number of fixes from 9.3.8.
14+
For information about new features in the 9.3 major release, see
15+
<xref linkend="release-9-3">.
16+
</para>
17+
18+
<sect2>
19+
<title>Migration to Version 9.3.9</title>
20+
21+
<para>
22+
A dump/restore is not required for those running 9.3.X.
23+
</para>
24+
25+
<para>
26+
However, if you are upgrading an installation that was previously
27+
upgraded using a <application>pg_upgrade</> version between 9.3.0 and
28+
9.3.4 inclusive, see the first changelog entry below.
29+
</para>
30+
31+
<para>
32+
Also, if you are upgrading from a version earlier than 9.3.7,
33+
see <xref linkend="release-9-3-7">.
34+
</para>
35+
36+
</sect2>
37+
38+
<sect2>
39+
<title>Changes</title>
40+
41+
<itemizedlist>
42+
43+
<listitem>
44+
<para>
45+
Fix possible failure to recover from an inconsistent database state
46+
(Robert Haas)
47+
</para>
48+
49+
<para>
50+
Recent <productname>PostgreSQL</> releases introduced mechanisms to
51+
protect against multixact wraparound, but some of that code did not
52+
account for the possibility that it would need to run during crash
53+
recovery, when the database may not be in a consistent state. This
54+
could result in failure to restart after a crash, or failure to start
55+
up a secondary server. The lingering effects of a previously-fixed
56+
bug in <application>pg_upgrade</> could also cause such a failure, in
57+
installations that had used <application>pg_upgrade</> versions
58+
between 9.3.0 and 9.3.4.
59+
</para>
60+
61+
<para>
62+
The <application>pg_upgrade</> bug in question was that it would
63+
set <literal>oldestMultiXid</> to 1 in <filename>pg_control</> even
64+
if the true value should be higher. With the fixes introduced in
65+
this release, such a situation will result in immediate emergency
66+
autovacuuming until a correct <literal>oldestMultiXid</> value can be
67+
determined. If that would pose a hardship, users can avoid it by
68+
doing manual vacuuming <emphasis>before</> upgrading to this release.
69+
In detail:
70+
71+
<orderedlist>
72+
<listitem>
73+
<para>
74+
Check whether <application>pg_controldata</> reports <quote>Latest
75+
checkpoint's oldestMultiXid</> to be 1. If not, there's nothing
76+
to do.
77+
</para>
78+
</listitem>
79+
<listitem>
80+
<para>
81+
Look in <filename>PGDATA/pg_multixact/offsets</> to see if there's a
82+
file named <filename>0000</>. If there is, there's nothing to do.
83+
</para>
84+
</listitem>
85+
<listitem>
86+
<para>
87+
Otherwise, for each table that has
88+
<structname>pg_class</>.<structfield>relminmxid</> equal to 1,
89+
<command>VACUUM</> that table with
90+
both <xref linkend="guc-vacuum-multixact-freeze-min-age">
91+
and <xref linkend="guc-vacuum-multixact-freeze-table-age"> set to
92+
zero. (You can use the vacuum cost delay parameters described
93+
in <xref linkend="runtime-config-resource-vacuum-cost"> to reduce
94+
the performance consequences for concurrent sessions.) You must
95+
use <productname>PostgreSQL</> 9.3.5 or later to perform this step.
96+
</para>
97+
</listitem>
98+
</orderedlist>
99+
</para>
100+
</listitem>
101+
102+
<listitem>
103+
<para>
104+
Fix rare failure to invalidate relation cache init file (Tom Lane)
105+
</para>
106+
107+
<para>
108+
With just the wrong timing of concurrent activity, a <command>VACUUM
109+
FULL</> on a system catalog might fail to update the <quote>init file</>
110+
that's used to avoid cache-loading work for new sessions. This would
111+
result in later sessions being unable to access that catalog at all.
112+
This is a very ancient bug, but it's so hard to trigger that no
113+
reproducible case had been seen until recently.
114+
</para>
115+
</listitem>
116+
117+
<listitem>
118+
<para>
119+
Avoid deadlock between incoming sessions and <literal>CREATE/DROP
120+
DATABASE</> (Tom Lane)
121+
</para>
122+
123+
<para>
124+
A new session starting in a database that is the target of
125+
a <command>DROP DATABASE</> command, or is the template for
126+
a <command>CREATE DATABASE</> command, could cause the command to wait
127+
for five seconds and then fail, even if the new session would have
128+
exited before that.
129+
</para>
130+
</listitem>
131+
132+
<listitem>
133+
<para>
134+
Improve planner's cost estimates for semi-joins and anti-joins with
135+
inner indexscans (Tom Lane, Tomas Vondra)
136+
</para>
137+
138+
<para>
139+
This type of plan is quite cheap when all the join clauses are used
140+
as index scan conditions, even if the inner scan would nominally
141+
fetch many rows, because the executor will stop after obtaining one
142+
row. The planner only partially accounted for that effect, and would
143+
therefore overestimate the cost, leading it to possibly choose some
144+
other much less efficient plan type.
145+
</para>
146+
</listitem>
147+
148+
</itemizedlist>
149+
150+
</sect2>
151+
</sect1>
152+
4153
<sect1 id="release-9-3-8">
5154
<title>Release 9.3.8</title>
6155

0 commit comments

Comments
 (0)