1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
|
<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><html xmlns="http://www.w3.org/1999/xhtml"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /><title>E.3. Release 15.2</title><link rel="stylesheet" type="text/css" href="stylesheet.css" /><link rev="made" href="pgsql-docs@lists.postgresql.org" /><meta name="generator" content="DocBook XSL Stylesheets Vsnapshot" /><link rel="prev" href="release-15-3.html" title="E.2. Release 15.3" /><link rel="next" href="release-15-1.html" title="E.4. Release 15.1" /></head><body id="docContent" class="container-fluid col-10"><div class="navheader"><table width="100%" summary="Navigation header"><tr><th colspan="5" align="center">E.3. Release 15.2</th></tr><tr><td width="10%" align="left"><a accesskey="p" href="release-15-3.html" title="E.2. Release 15.3">Prev</a> </td><td width="10%" align="left"><a accesskey="u" href="release.html" title="Appendix E. Release Notes">Up</a></td><th width="60%" align="center">Appendix E. Release Notes</th><td width="10%" align="right"><a accesskey="h" href="index.html" title="PostgreSQL 15.4 Documentation">Home</a></td><td width="10%" align="right"> <a accesskey="n" href="release-15-1.html" title="E.4. Release 15.1">Next</a></td></tr></table><hr /></div><div class="sect1" id="RELEASE-15-2"><div class="titlepage"><div><div><h2 class="title" style="clear: both">E.3. Release 15.2</h2></div></div></div><div class="toc"><dl class="toc"><dt><span class="sect2"><a href="release-15-2.html#id-1.11.6.7.4">E.3.1. Migration to Version 15.2</a></span></dt><dt><span class="sect2"><a href="release-15-2.html#id-1.11.6.7.5">E.3.2. Changes</a></span></dt></dl></div><p><strong>Release date: </strong>2023-02-09</p><p>
This release contains a variety of fixes from 15.1.
For information about new features in major release 15, see
<a class="xref" href="release-15.html" title="E.5. Release 15">Section E.5</a>.
</p><div class="sect2" id="id-1.11.6.7.4"><div class="titlepage"><div><div><h3 class="title">E.3.1. Migration to Version 15.2</h3></div></div></div><p>
A dump/restore is not required for those running 15.X.
</p><p>
However, if you are upgrading from a version earlier than 15.1,
see <a class="xref" href="release-15-1.html" title="E.4. Release 15.1">Section E.4</a>.
</p></div><div class="sect2" id="id-1.11.6.7.5"><div class="titlepage"><div><div><h3 class="title">E.3.2. Changes</h3></div></div></div><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem"><p>
<span class="application">libpq</span> can leak memory contents after
GSSAPI transport encryption initiation fails (Jacob Champion)
</p><p>
A modified server, or an unauthenticated man-in-the-middle, can
send a not-zero-terminated error message during setup of GSSAPI
(Kerberos) transport encryption. <span class="application">libpq</span>
will then copy that string, as well as following bytes in
application memory up to the next zero byte, to its error report.
Depending on what the calling application does with the error
report, this could result in disclosure of application memory
contents. There is also a small probability of a crash due to
reading beyond the end of memory. Fix by properly zero-terminating
the server message.
(CVE-2022-41862)
</p></li><li class="listitem"><p>
Fix calculation of which <code class="literal">GENERATED</code> columns need
to be updated in child tables during an <code class="command">UPDATE</code> on
a partitioned table or inheritance tree (Amit Langote, Tom Lane)
</p><p>
This fixes failure to update <code class="literal">GENERATED</code> columns
that do not exist in the parent table, or that have different
dependencies than are in the parent column's generation expression.
</p></li><li class="listitem"><p>
Fix possible failure of <code class="command">MERGE</code> to compute
<code class="literal">GENERATED</code> columns (Dean Rasheed)
</p><p>
When the first row-level action of the <code class="command">MERGE</code> was
an <code class="literal">UPDATE</code>, any
subsequent <code class="literal">INSERT</code> actions would fail to
compute <code class="literal">GENERATED</code> columns that were deemed
unnecessary to compute for the <code class="literal">UPDATE</code> action
(due to not depending on any of the <code class="literal">UPDATE</code> target
columns).
</p></li><li class="listitem"><p>
Fix <code class="command">MERGE</code>'s check for
unreachable <code class="literal">WHEN</code> clauses (Dean Rasheed)
</p><p>
A <code class="literal">WHEN</code> clause following an
unconditional <code class="literal">WHEN</code> clause should be rejected as
unreachable, but this case was not always detected.
</p></li><li class="listitem"><p>
Fix <code class="command">MERGE</code>'s rule-detection test (Dean Rasheed)
</p><p>
<code class="command">MERGE</code> is not supported on tables with rules;
but it also failed on tables that once had rules but no longer do.
</p></li><li class="listitem"><p>
In <code class="command">MERGE</code>, don't count a <code class="literal">DO
NOTHING</code> action as a processed tuple (Álvaro Herrera)
</p><p>
This makes the code's behavior match the documentation.
</p></li><li class="listitem"><p>
Allow a <code class="literal">WITH RECURSIVE ... CYCLE</code> CTE
to access its output column (Tom Lane)
</p><p>
A reference to the <code class="literal">SET</code> column from within the CTE
would fail with <span class="quote">“<span class="quote">cache lookup failed for type 0</span>”</span>.
</p></li><li class="listitem"><p>
Fix handling of pending inserts when doing a bulk insertion to a
foreign table (Etsuro Fujita)
</p><p>
In some cases pending insertions were not flushed to the FDW soon
enough, leading to logical inconsistencies, for
example <code class="literal">BEFORE ROW</code> triggers not seeing rows they
should be able to see.
</p></li><li class="listitem"><p>
Allow <code class="literal">REPLICA IDENTITY</code>
to be set on an index that's not (yet) valid (Tom Lane)
</p><p>
When <span class="application">pg_dump</span> dumps a partitioned index
that's marked <code class="literal">REPLICA IDENTITY</code>, it generates a
command sequence that applies <code class="literal">REPLICA IDENTITY</code>
before the partitioned index has been marked valid, causing restore
to fail. There seems no very good reason to prohibit doing it in
that order, so allow it. The marking will have no effect anyway
until the index becomes valid.
</p></li><li class="listitem"><p>
Fix handling of <code class="literal">DEFAULT</code> markers in rules that
perform an <code class="command">INSERT</code> from a
multi-row <code class="literal">VALUES</code> list (Dean Rasheed)
</p><p>
In some cases a <code class="literal">DEFAULT</code> marker would not get
replaced with the proper default-value expression, leading to
an <span class="quote">“<span class="quote">unrecognized node type</span>”</span> error.
</p></li><li class="listitem"><p>
Reject uses of undefined variables in <code class="type">jsonpath</code>
existence checks (Alexander Korotkov, David G. Johnston)
</p><p>
While <code class="type">jsonpath</code> match operators threw an error for an
undefined variable in the path pattern, the existence operators
silently treated it as a match.
</p></li><li class="listitem"><p>
Fix <code class="type">jsonb</code> subscripting to cope with toasted subscript
values (Tom Lane, David G. Johnston)
</p><p>
Using a text value fetched directly from a table as
a <code class="type">jsonb</code> subscript was likely to fail.
Fetches would usually not find any matching element.
Assignments could store the value with a garbage key,
although keys long enough to cause that problem are probably rare in
the field.
</p></li><li class="listitem"><p>
Fix edge-case data corruption in parallel hash joins (Dmitry Astapov)
</p><p>
If the final chunk of a large tuple being written out to a temporary
file was exactly 32760 bytes, it would be corrupted due to a
fencepost bug. The query would typically fail later with
corrupted-data symptoms.
</p></li><li class="listitem"><p>
Honor non-default settings
of <code class="varname">checkpoint_completion_target</code>
(Bharath Rupireddy)
</p><p>
Internal state was not updated after a change
in <code class="varname">checkpoint_completion_target</code>, possibly
resulting in performing checkpoint I/O faster or slower than
desired, especially if that setting was changed on-the-fly.
</p></li><li class="listitem"><p>
Log the correct ending timestamp
in <code class="varname">recovery_target_xid</code> mode (Tom Lane)
</p><p>
When ending recovery based on the <code class="varname">recovery_target_xid</code>
setting with <code class="varname">recovery_target_inclusive</code>
= <code class="literal">off</code>, we printed an incorrect timestamp (always
2000-01-01) in the <span class="quote">“<span class="quote">recovery stopping before
... transaction</span>”</span> log message.
</p></li><li class="listitem"><p>
Improve error reporting for some buffered file read failures
(Peter Eisentraut)
</p><p>
Correctly report a short read, giving the numbers of bytes desired
and actually read, instead of reporting an irrelevant error code.
Most places got this right already, but some recently-written
replication logic did not.
</p></li><li class="listitem"><p>
Remove arbitrary limit on number of elements
in <code class="type">int2vector</code> and <code class="type">oidvector</code> (Tom Lane)
</p><p>
The input functions for these types previously rejected more than
100 elements. With the introduction of the logical replication
column list feature, it's necessary to
accept <code class="type">int2vector</code>s having up to 1600 columns,
otherwise long column lists cause logical-replication failures.
</p></li><li class="listitem"><p>
In extended query protocol, avoid an immediate commit
after <code class="command">ANALYZE</code> if we're running a pipeline
(Tom Lane)
</p><p>
If there's not been an explicit <code class="command">BEGIN
TRANSACTION</code>, <code class="command">ANALYZE</code> would take it on
itself to commit, which should not happen within a pipelined series
of commands.
</p></li><li class="listitem"><p>
Reject cancel request packets having the wrong length
(Andrey Borodin)
</p><p>
The server would process a cancel request even if its length word
was too small. This led to reading beyond the end of the allocated
buffer. In theory that could cause a segfault, but it seems quite
unlikely to happen in practice, since the buffer would have to be
very close to the end of memory. The more likely outcome was a bogus
log message about wrong backend PID or cancel code. Complain about
the wrong length, instead.
</p></li><li class="listitem"><p>
Fix planner preprocessing oversights for window function run-condition
expressions (Richard Guo, David Rowley)
</p><p>
This could lead to planner errors such as <span class="quote">“<span class="quote">WindowFunc not
found in subplan target lists</span>”</span>.
</p></li><li class="listitem"><p>
Fix possible dangling-pointer access during execution of window
function run-condition expressions (David Rowley)
</p><p>
In practice, because the run-condition optimization is only applied
to certain window functions that happen to all
return <code class="type">int8</code>, this only manifested as a problem on
32-bit builds.
</p></li><li class="listitem"><p>
Add recursion and looping defenses in subquery pullup (Tom Lane)
</p><p>
A contrived query can result in deep recursion and unreasonable
amounts of time spent trying to flatten subqueries. A proper fix
for that seems unduly invasive for a back-patch, but we can at least
add stack depth checks and an interrupt check to allow the query to
be cancelled.
</p></li><li class="listitem"><p>
Fix planner issues when combining Memoize nodes with partitionwise
joins or parameterized nestloops (Richard Guo)
</p><p>
These errors could lead to not using Memoize in contexts where it
would be useful, or possibly to wrong query plans.
</p></li><li class="listitem"><p>
Fix partitionwise-join code to tolerate failure to produce a plan for
each partition (Tom Lane)
</p><p>
This could result in <span class="quote">“<span class="quote">could not devise a query plan for the
given query</span>”</span> errors.
</p></li><li class="listitem"><p>
Limit the amount of cleanup work done
by <code class="function">get_actual_variable_range</code> (Simon Riggs)
</p><p>
Planner runs occurring just after deletion of a large number of
tuples appearing at the end of an index could expend significant
amounts of work setting the <span class="quote">“<span class="quote">killed</span>”</span> bits for those
index entries. Limit the amount of work done in any one query by
giving up on this process after examining 100 heap pages. All the
cleanup will still happen eventually, but without so large a
performance hiccup.
</p></li><li class="listitem"><p>
Prevent the statistics machinery from getting confused when a
relation's relkind changes (Andres Freund)
</p><p>
Converting a table to a view could lead to crashes or assertion
failures.
</p></li><li class="listitem"><p>
Fix under-parenthesized display of <code class="literal">AT TIME ZONE</code>
constructs (Tom Lane)
</p><p>
This could result in dump/restore failures for rules or views in
which an argument of <code class="literal">AT TIME ZONE</code> is itself an
expression.
</p></li><li class="listitem"><p>
Prevent clobbering of cached parsetrees for utility statements in
SQL functions (Tom Lane, Daniel Gustafsson)
</p><p>
If a SQL-language function executes the same utility command more
than once within a single calling query, it could crash or report
strange errors such as <span class="quote">“<span class="quote">unrecognized node type</span>”</span>.
</p></li><li class="listitem"><p>
Ensure that execution of full-text-search queries can be cancelled
while they are performing phrase matches (Tom Lane)
</p></li><li class="listitem"><p>
Fix memory leak in hashing strings with nondeterministic collations
(Jeff Davis)
</p></li><li class="listitem"><p>
Fix deadlock between <code class="command">DROP DATABASE</code> and logical
replication worker process (Hou Zhijie)
</p><p>
This was caused by an ill-advised choice to block interrupts while
creating a logical replication slot in the worker. In version 15
that could lead to an undetected deadlock. In version 14, no
deadlock has been observed, but it's still a bad idea to block
interrupts while waiting for network I/O.
</p></li><li class="listitem"><p>
Clean up the <span class="application">libpq</span> connection object
after a failed replication connection attempt (Andres Freund)
</p><p>
The previous coding leaked the connection object. In background
code paths that's pretty harmless because the calling process will
give up and exit. But in commands such as <code class="command">CREATE
SUBSCRIPTION</code>, such a failure resulted in a small
session-lifespan memory leak.
</p></li><li class="listitem"><p>
In hot-standby servers, reduce processing effort for tracking XIDs
known to be active on the primary (Simon Riggs, Michail Nikolaev)
</p><p>
Insufficiently-aggressive cleanup of the KnownAssignedXids array
could lead to poor performance, particularly
when <code class="varname">max_connections</code> is set to a large value on
the standby.
</p></li><li class="listitem"><p>
Ignore invalidated logical-replication slots while determining
oldest catalog xmin (Sirisha Chamarthi)
</p><p>
A replication slot could prevent cleanup of dead tuples in the
system catalogs even after it becomes invalidated due to
exceeding <code class="varname">max_slot_wal_keep_size</code>. Thus, failure
of a replication consumer could lead to indefinitely-large catalog
bloat.
</p></li><li class="listitem"><p>
In logical decoding, notify the remote node when a transaction is
detected to have crashed (Hou Zhijie)
</p><p>
After a server restart, we'll re-stream the changes for transactions
occurring shortly before the restart. Some of these transactions
probably never completed; when we realize that one didn't we throw
away the relevant decoding state locally, but we neglected to tell
the subscriber about it. That led to the subscriber keeping useless
streaming files until it's next restarted.
</p></li><li class="listitem"><p>
Fix uninitialized-memory usage in logical decoding (Masahiko Sawada)
</p><p>
In certain cases, resumption of logical decoding could try to re-use
XID data that had already been freed, leading to unpredictable
behavior.
</p></li><li class="listitem"><p>
Acquire spinlock while updating shared state during logical decoding
context creation (Masahiko Sawada)
</p><p>
We neglected to acquire the appropriate lock while updating data
about two-phase transactions, potentially allowing other processes
to see inconsistent data.
</p></li><li class="listitem"><p>
Fix <span class="application">pgoutput</span> replication plug-in to not
send columns not listed in a table's replication column list
(Hou Zhijie)
</p><p>
<code class="literal">UPDATE</code> and <code class="literal">DELETE</code> events did
not pay attention to the configured column list, thus sending more
data than expected. This did not cause a problem when the receiver
is our built-in logical replication code, but it might confuse other
receivers, and in any case it wasted network bandwidth.
</p></li><li class="listitem"><p>
Avoid rare <span class="quote">“<span class="quote">failed to acquire cleanup lock</span>”</span> panic
during WAL replay of hash-index page split operations (Robert Haas)
</p></li><li class="listitem"><p>
Advance a heap page's LSN when setting its all-visible bit during
WAL replay (Jeff Davis)
</p><p>
Failure to do this left the page possibly different on standby
servers than the primary, and violated some other expectations about
when the LSN changes. This seems only a theoretical hazard so
far as <span class="productname">PostgreSQL</span> itself is concerned,
but it could upset third-party tools.
</p></li><li class="listitem"><p>
Fix <code class="function">int64_div_fast_to_numeric()</code> to work for a
wider range of inputs (Dean Rasheed)
</p><p>
This function misbehaved with some values of its second argument.
No such usages exist in core <span class="productname">PostgreSQL</span>,
but it's clearly a hazard for external modules, so repair.
</p></li><li class="listitem"><p>
Fix latent buffer-overrun problem in <code class="literal">WaitEventSet</code>
logic (Thomas Munro)
</p><p>
The <code class="function">epoll</code>-based
and <code class="function">kqueue</code>-based implementations could ask the
kernel for too many events if the size of their internal buffer was
different from the size of the caller's output buffer. That case is
not known to occur in released <span class="productname">PostgreSQL</span>
versions, but this error is a hazard for external modules and future
bug fixes.
</p></li><li class="listitem"><p>
Avoid nominally-undefined behavior when accessing shared memory in
32-bit builds (Andres Freund)
</p><p>
clang's undefined-behavior sanitizer complained about use of a
pointer that was less aligned than it should be. It's very unlikely
that this would cause a problem in non-debug builds, but it's worth
fixing for testing purposes.
</p></li><li class="listitem"><p>
Fix assertion failure in BRIN minmax-multi opclasses (Tomas Vondra)
</p><p>
The assertion was overly strict, so this mistake was harmless in
non-assert builds.
</p></li><li class="listitem"><p>
Remove faulty assertion in useless-RESULT-RTE optimization logic
(Tom Lane)
</p></li><li class="listitem"><p>
Fix copy-and-paste errors in cache-lookup-failure messages for ACL
checks (Justin Pryzby)
</p><p>
In principle these errors should never be reached. But if they are,
some of them reported the wrong type of object.
</p></li><li class="listitem"><p>
Fix possible corruption of very large tablespace map files
in <span class="application">pg_basebackup</span> (Antonin Houska)
</p></li><li class="listitem"><p>
Avoid harmless warning from <span class="application">pg_dump</span>
in <code class="option">--if-exists</code> mode (Tom Lane)
</p><p>
If the <code class="literal">public</code> schema has a non-default owner then
use of <span class="application">pg_dump</span>'s <code class="option">--if-exists</code>
option resulted in a warning message <span class="quote">“<span class="quote">warning: could not find
where to insert IF EXISTS in statement "-- *not* dropping schema,
since initdb creates it"</span>”</span>. The dump output was okay, though.
</p></li><li class="listitem"><p>
Fix <span class="application">psql</span>'s <code class="literal">\sf</code>
and <code class="literal">\ef</code> commands to handle SQL-language functions
that have <acronym class="acronym">SQL</acronym>-standard function bodies (Tom Lane)
</p><p>
These commands misidentified the start of the function body when it
used new-style syntax.
</p></li><li class="listitem"><p>
Fix tab completion of <code class="command">ALTER
FUNCTION/PROCEDURE/ROUTINE</code> ... <code class="command">SET
SCHEMA</code> (Dean Rasheed)
</p></li><li class="listitem"><p>
Update <code class="filename">contrib/pageinspect</code> to mark its
disk-accessing functions as <code class="literal">PARALLEL RESTRICTED</code>
(Tom Lane)
</p><p>
This avoids possible failure if one of these functions is used to
examine a temporary table, since a session's temporary tables are not
accessible from parallel workers.
</p></li><li class="listitem"><p>
Fix <code class="filename">contrib/seg</code> to not crash or print garbage
if an input number has more than 127 digits (Tom Lane)
</p></li><li class="listitem"><p>
Fix build on Microsoft Visual Studio 2013 (Tom Lane)
</p><p>
A previous patch supposed that all platforms of interest
have <code class="function">snprintf()</code>, but MSVC 2013 isn't quite
there yet. Revert to using <code class="function">sprintf()</code> on that
platform.
</p></li><li class="listitem"><p>
Fix compile failure in building PL/Perl with MSVC when using
Strawberry Perl (Andrew Dunstan)
</p></li><li class="listitem"><p>
Fix mismatch of PL/Perl built with MSVC versus a Perl library built
with gcc (Andrew Dunstan)
</p><p>
Such combinations could previously fail with <span class="quote">“<span class="quote">loadable library
and perl binaries are mismatched</span>”</span> errors.
</p></li><li class="listitem"><p>
Suppress compiler warnings from Perl's header files (Andres Freund)
</p><p>
Our preferred compiler options provoke warnings about constructs
appearing in recent versions of Perl's header files. When using
<span class="application">gcc</span>, we can suppress these warnings with
a pragma.
</p></li><li class="listitem"><p>
Fix <span class="application">pg_waldump</span> to build on compilers that
don't discard unused static-inline functions (Tom Lane)
</p></li><li class="listitem"><p>
Update time zone data files to <span class="application">tzdata</span>
release 2022g for DST law changes in Greenland and Mexico,
plus historical corrections for northern Canada, Colombia, and
Singapore.
</p><p>
Notably, a new timezone America/Ciudad_Juarez has been split off
from America/Ojinaga.
</p></li></ul></div></div></div><div class="navfooter"><hr /><table width="100%" summary="Navigation footer"><tr><td width="40%" align="left"><a accesskey="p" href="release-15-3.html" title="E.2. Release 15.3">Prev</a> </td><td width="20%" align="center"><a accesskey="u" href="release.html" title="Appendix E. Release Notes">Up</a></td><td width="40%" align="right"> <a accesskey="n" href="release-15-1.html" title="E.4. Release 15.1">Next</a></td></tr><tr><td width="40%" align="left" valign="top">E.2. Release 15.3 </td><td width="20%" align="center"><a accesskey="h" href="index.html" title="PostgreSQL 15.4 Documentation">Home</a></td><td width="40%" align="right" valign="top"> E.4. Release 15.1</td></tr></table></div></body></html>
|
