1
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
8
CONTENT="Modular DocBook HTML Stylesheet Version 1.79"><LINK
10
HREF="mailto:pgsql-docs@postgresql.org"><LINK
12
TITLE="PostgreSQL 9.1beta1 Documentation"
13
HREF="index.html"><LINK
16
HREF="release.html"><LINK
19
HREF="release-8-3-2.html"><LINK
22
HREF="release-8-3.html"><LINK
25
HREF="stylesheet.css"><META
26
HTTP-EQUIV="Content-Type"
27
CONTENT="text/html; charset=ISO-8859-1"><META
29
CONTENT="2011-04-27T21:20:33"></HEAD
35
SUMMARY="Header navigation table"
47
>PostgreSQL 9.1beta1 Documentation</A
57
HREF="release-8-3-2.html"
74
>Appendix E. Release Notes</TD
90
HREF="release-8-3.html"
105
>E.30. Release 8.3.1</A
118
> This release contains a variety of fixes from 8.3.0.
119
For information about new features in the 8.3 major release, see
121
HREF="release-8-3.html"
131
>E.30.1. Migration to Version 8.3.1</A
134
> A dump/restore is not required for those running 8.3.X.
135
However, you might need to <TT
139
columns after updating, if you are affected by the Windows locale
140
issue described below.
156
> Fix character string comparison for Windows locales that consider
157
different character combinations as equal (Tom)
160
> This fix applies only on Windows and only when using UTF-8
161
database encoding. The same fix was made for all other cases
162
over two years ago, but Windows with UTF-8 uses a separate code
163
path that was not updated. If you are using a locale that
164
considers some non-identical strings as equal, you may need to
168
> to fix existing indexes on textual columns.
173
> Repair corner-case bugs in <TT
179
> A potential deadlock between concurrent <TT
183
operations on different system catalogs was introduced in 8.2.
184
This has now been corrected. 8.3 made this worse because the
185
deadlock could occur within a critical code section, making it
186
a PANIC rather than just ERROR condition.
192
> that failed partway through
193
vacuuming a system catalog could result in cache corruption in
194
concurrent database sessions.
200
> bug introduced in 8.3 could
201
result in a crash or out-of-memory report when dealing with
202
pages containing no live tuples.
207
> Fix misbehavior of foreign key checks involving <TT
217
> If the referencing column were of a different but compatible type
221
>), the constraint was enforced incorrectly.
226
> Avoid needless deadlock failures in no-op foreign-key checks (Stephan
232
> Fix possible core dump when re-planning a prepared query (Tom)
235
> This bug affected only protocol-level prepare operations, not
239
>, and so tended to be seen only with
240
JDBC, DBI, and other client-side drivers that use prepared
246
> Fix possible failure when re-planning a query that calls an SPI-using
252
> Fix failure in row-wise comparisons involving columns of different
258
> Fix longstanding <TT
268
> In rare cases a session that had just executed a
272
> might not get a notification, even though
273
one would be expected because the concurrent transaction executing
277
> was observed to commit later.
280
> A side effect of the fix is that a transaction that has executed
281
a not-yet-committed <TT
284
> command will not see any
292
should it choose to look; formerly it would have. This behavior
293
was never documented one way or the other, but it is possible that
294
some applications depend on the old behavior.
306
prepared transaction (Tom)
309
> This was formerly allowed but trying to do it had various unpleasant
310
consequences, notably that the originating backend could not exit
314
> remained uncommitted.
319
> Disallow dropping a temporary table within a
320
prepared transaction (Heikki)
323
> This was correctly disallowed by 8.1, but the check was inadvertently
324
broken in 8.2 and 8.3.
329
> Fix rare crash when an error occurs during a query using a hash index
335
> Fix incorrect comparison of <TT
343
> Fix incorrect behavior of <TT
346
> with non-ASCII characters
347
in single-byte encodings (Rolf Jentsch)
358
> This function should have been removed before 8.3 release, but
359
was inadvertently left in the source code. It poses a small
360
security risk since unprivileged users could use it to read the
361
first few characters of any file accessible to the server.
366
> Fix memory leaks in certain usages of set-returning functions (Neil)
380
high-bit-set byte values into <TT
389
escape sequences (Tom)
392
> This is necessary to avoid encoding problems when the database
393
encoding is multi-byte. This change could pose compatibility issues
394
for applications that are expecting specific results from
403
> Fix input of datetime values for February 29 in years BC (Tom)
406
> The former coding was mistaken about which years were leap years.
413
>"unrecognized node type"</SPAN
414
> error in some variants of
423
> Avoid tablespace permissions errors in <TT
426
INCLUDING INDEXES</TT
434
>pg_stat_activity</TT
439
is cleared when a lock wait is aborted (Tom)
444
> Fix handling of process permissions on Windows Vista (Dave, Magnus)
447
> In particular, this fix allows starting the server as the Administrator
453
> Update time zone data files to <SPAN
457
(in particular, recent Chile changes); adjust timezone abbreviation
461
> (Venezuela) to mean UTC-4:30, not UTC-4:00 (Tom)
469
> problems with arrays (Michael)
477
> to correctly extract the postmaster's port
478
number from command-line options (Itagaki Takahiro, Tom)
484
> could try to contact the
485
postmaster on the wrong port, leading to bogus reports of startup
494
> to defend against possible misoptimization
501
> This is known to be necessary when building <SPAN
513
> Enable building <TT
515
>contrib/uuid-ossp</TT
516
> with MSVC (Hiroshi Saito)
527
SUMMARY="Footer navigation table"
538
HREF="release-8-3-2.html"
556
HREF="release-8-3.html"
b'\\ No newline at end of file'