pgsql: hash: Add write-ahead logging support.

pg.ddx.io  pgsql-committers@postgresql.org mailing list archive  
help / color / mirror / Atom feed

pgsql: hash: Add write-ahead logging support.
13+ messages / 5 participants
[nested] [flat]

* pgsql: hash: Add write-ahead logging support.
@ 2017-03-14 17:27  Robert Haas <rhaas@postgresql.org>
  0 siblings, 1 reply; 13+ messages in thread

From: Robert Haas @ 2017-03-14 17:27 UTC (permalink / raw)
  To: pgsql-committers

hash: Add write-ahead logging support.

The warning about hash indexes not being write-ahead logged and their
use being discouraged has been removed.  "snapshot too old" is now
supported for tables with hash indexes.  Most importantly, barring
bugs, hash indexes will now be crash-safe and usable on standbys.

This commit doesn't yet add WAL consistency checking for hash
indexes, as we now have for other index types; a separate patch has
been submitted to cure that lack.

Amit Kapila, reviewed and slightly modified by me.  The larger patch
series of which this is a part has been reviewed and tested by Álvaro
Herrera, Ashutosh Sharma, Mark Kirkwood, Jeff Janes, and Jesper
Pedersen.

Discussion: http://postgr.es/m/CAA4eK1JOBX=YU33631Qh-XivYXtPSALh514+jR8XeD7v+K3r_Q@mail.gmail.com

Branch
------
master

Details
-------
http://git.postgresql.org/pg/commitdiff/c11453ce0aeaa377cbbcc9a3fc418acb94629330

Modified Files
--------------
contrib/pageinspect/expected/hash.out          |   1 -
contrib/pgstattuple/expected/pgstattuple.out   |   2 -
doc/src/sgml/backup.sgml                       |  13 -
doc/src/sgml/config.sgml                       |   7 +-
doc/src/sgml/high-availability.sgml            |   6 -
doc/src/sgml/indices.sgml                      |  12 -
doc/src/sgml/ref/create_index.sgml             |  13 -
src/backend/access/hash/Makefile               |   2 +-
src/backend/access/hash/README                 | 138 +++-
src/backend/access/hash/hash.c                 |  81 ++-
src/backend/access/hash/hash_xlog.c            | 963 +++++++++++++++++++++++++
src/backend/access/hash/hashinsert.c           |  59 +-
src/backend/access/hash/hashovfl.c             | 209 +++++-
src/backend/access/hash/hashpage.c             | 236 +++++-
src/backend/access/hash/hashsearch.c           |   5 +
src/backend/access/rmgrdesc/hashdesc.c         | 134 +++-
src/backend/commands/indexcmds.c               |   5 -
src/backend/utils/cache/relcache.c             |  12 +-
src/include/access/hash_xlog.h                 | 232 ++++++
src/test/regress/expected/create_index.out     |   5 -
src/test/regress/expected/enum.out             |   1 -
src/test/regress/expected/hash_index.out       |   2 -
src/test/regress/expected/macaddr.out          |   1 -
src/test/regress/expected/replica_identity.out |   1 -
src/test/regress/expected/uuid.out             |   1 -
25 files changed, 2001 insertions(+), 140 deletions(-)


-- 
Sent via pgsql-committers mailing list (pgsql-committers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-committers


^ permalink  raw  reply  [nested|flat] 13+ messages in thread

* Re: pgsql: hash: Add write-ahead logging support.
@ 2017-03-14 17:54  Tom Lane <tgl@sss.pgh.pa.us>
  parent: Robert Haas <rhaas@postgresql.org>
  0 siblings, 1 reply; 13+ messages in thread

From: Tom Lane @ 2017-03-14 17:54 UTC (permalink / raw)
  To: Robert Haas <rhaas@postgresql.org>; +Cc: pgsql-committers

Robert Haas <rhaas@postgresql.org> writes:
> hash: Add write-ahead logging support.

This bit in access/hash/README, lines 368ff, appears obsolete:

  Although we can survive a failure to split a bucket, a crash is likely to
  corrupt the index, since hash indexes are not yet WAL-logged.

Perhaps we can just delete that para.  If not, what should it say now?

			regards, tom lane

-- 
Sent via pgsql-committers mailing list (pgsql-committers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-committers

^ permalink  raw  reply  [nested|flat] 13+ messages in thread

* Re: pgsql: hash: Add write-ahead logging support.
@ 2017-03-14 17:56  Robert Haas <robertmhaas@gmail.com>
  parent: Tom Lane <tgl@sss.pgh.pa.us>
  0 siblings, 2 replies; 13+ messages in thread

From: Robert Haas @ 2017-03-14 17:56 UTC (permalink / raw)
  To: Tom Lane <tgl@sss.pgh.pa.us>; +Cc: pgsql-committers

On Tue, Mar 14, 2017 at 1:54 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Robert Haas <rhaas@postgresql.org> writes:
>> hash: Add write-ahead logging support.
>
> This bit in access/hash/README, lines 368ff, appears obsolete:
>
>   Although we can survive a failure to split a bucket, a crash is likely to
>   corrupt the index, since hash indexes are not yet WAL-logged.
>
> Perhaps we can just delete that para.  If not, what should it say now?

That's obsolete, and can just be removed.

-- 
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company


-- 
Sent via pgsql-committers mailing list (pgsql-committers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-committers



^ permalink  raw  reply  [nested|flat] 13+ messages in thread

* Re: pgsql: hash: Add write-ahead logging support.
@ 2017-03-14 18:18  Tom Lane <tgl@sss.pgh.pa.us>
  parent: Robert Haas <robertmhaas@gmail.com>
  1 sibling, 1 reply; 13+ messages in thread

From: Tom Lane @ 2017-03-14 18:18 UTC (permalink / raw)
  To: Robert Haas <robertmhaas@gmail.com>; +Cc: pgsql-committers

Robert Haas <robertmhaas@gmail.com> writes:
> On Tue, Mar 14, 2017 at 1:54 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> Robert Haas <rhaas@postgresql.org> writes:
>>> hash: Add write-ahead logging support.

>> This bit in access/hash/README, lines 368ff, appears obsolete:
>> 
>> Although we can survive a failure to split a bucket, a crash is likely to
>> corrupt the index, since hash indexes are not yet WAL-logged.
>> 
>> Perhaps we can just delete that para.  If not, what should it say now?

> That's obsolete, and can just be removed.

You gonna do it, or you want me to?

			regards, tom lane


-- 
Sent via pgsql-committers mailing list (pgsql-committers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-committers



^ permalink  raw  reply  [nested|flat] 13+ messages in thread

* Re: pgsql: hash: Add write-ahead logging support.
@ 2017-03-14 18:22  Robert Haas <robertmhaas@gmail.com>
  parent: Tom Lane <tgl@sss.pgh.pa.us>
  0 siblings, 1 reply; 13+ messages in thread

From: Robert Haas @ 2017-03-14 18:22 UTC (permalink / raw)
  To: Tom Lane <tgl@sss.pgh.pa.us>; +Cc: pgsql-committers

On Tue, Mar 14, 2017 at 2:18 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Robert Haas <robertmhaas@gmail.com> writes:
>> On Tue, Mar 14, 2017 at 1:54 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>>> Robert Haas <rhaas@postgresql.org> writes:
>>>> hash: Add write-ahead logging support.
>
>>> This bit in access/hash/README, lines 368ff, appears obsolete:
>>>
>>> Although we can survive a failure to split a bucket, a crash is likely to
>>> corrupt the index, since hash indexes are not yet WAL-logged.
>>>
>>> Perhaps we can just delete that para.  If not, what should it say now?
>
>> That's obsolete, and can just be removed.
>
> You gonna do it, or you want me to?

If you're feeling motivated, have at it.  Otherwise, I'll do it in a
day or two after I see what other complaints get reported that might
be addressed at the same time.

-- 
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company


-- 
Sent via pgsql-committers mailing list (pgsql-committers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-committers



^ permalink  raw  reply  [nested|flat] 13+ messages in thread

* Re: pgsql: hash: Add write-ahead logging support.
@ 2017-03-14 18:39  Tom Lane <tgl@sss.pgh.pa.us>
  parent: Robert Haas <robertmhaas@gmail.com>
  0 siblings, 0 replies; 13+ messages in thread

From: Tom Lane @ 2017-03-14 18:39 UTC (permalink / raw)
  To: Robert Haas <robertmhaas@gmail.com>; +Cc: pgsql-committers

Robert Haas <robertmhaas@gmail.com> writes:
> On Tue, Mar 14, 2017 at 2:18 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> You gonna do it, or you want me to?

> If you're feeling motivated, have at it.  Otherwise, I'll do it in a
> day or two after I see what other complaints get reported that might
> be addressed at the same time.

Nah, long as it's on your to-do list that's good enough for me.

			regards, tom lane


-- 
Sent via pgsql-committers mailing list (pgsql-committers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-committers



^ permalink  raw  reply  [nested|flat] 13+ messages in thread

* Re: pgsql: hash: Add write-ahead logging support.
@ 2017-03-15 02:23  Amit Kapila <amit.kapila16@gmail.com>
  parent: Robert Haas <robertmhaas@gmail.com>
  1 sibling, 1 reply; 13+ messages in thread

From: Amit Kapila @ 2017-03-15 02:23 UTC (permalink / raw)
  To: Robert Haas <robertmhaas@gmail.com>; +Cc: Tom Lane <tgl@sss.pgh.pa.us>; pgsql-committers

On Tue, Mar 14, 2017 at 11:26 PM, Robert Haas <robertmhaas@gmail.com> wrote:
> On Tue, Mar 14, 2017 at 1:54 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> Robert Haas <rhaas@postgresql.org> writes:
>>> hash: Add write-ahead logging support.
>>
>> This bit in access/hash/README, lines 368ff, appears obsolete:
>>
>>   Although we can survive a failure to split a bucket, a crash is likely to
>>   corrupt the index, since hash indexes are not yet WAL-logged.
>>
>> Perhaps we can just delete that para.  If not, what should it say now?
>
> That's obsolete, and can just be removed.
>

Agreed and posted a patch on hackers [1]

[1] - https://www.postgresql.org/message-id/CAA4eK1LeVu8rSvA%3DhfbezqjNv5x3nU-rPbHCHHbbsCrLX-0zAg%40mail.g...

-- 
With Regards,
Amit Kapila.
EnterpriseDB: http://www.enterprisedb.com


-- 
Sent via pgsql-committers mailing list (pgsql-committers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-committers



^ permalink  raw  reply  [nested|flat] 13+ messages in thread

* Re: pgsql: hash: Add write-ahead logging support.
@ 2017-03-15 02:35  Thomas Munro <thomas.munro@enterprisedb.com>
  parent: Amit Kapila <amit.kapila16@gmail.com>
  0 siblings, 1 reply; 13+ messages in thread

From: Thomas Munro @ 2017-03-15 02:35 UTC (permalink / raw)
  To: Amit Kapila <amit.kapila16@gmail.com>; +Cc: Robert Haas <robertmhaas@gmail.com>; Tom Lane <tgl@sss.pgh.pa.us>; pgsql-committers

On Wed, Mar 15, 2017 at 3:23 PM, Amit Kapila <amit.kapila16@gmail.com> wrote:
> On Tue, Mar 14, 2017 at 11:26 PM, Robert Haas <robertmhaas@gmail.com> wrote:
>> On Tue, Mar 14, 2017 at 1:54 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>>> Robert Haas <rhaas@postgresql.org> writes:
>>>> hash: Add write-ahead logging support.
>>>
>>> This bit in access/hash/README, lines 368ff, appears obsolete:
>>>
>>>   Although we can survive a failure to split a bucket, a crash is likely to
>>>   corrupt the index, since hash indexes are not yet WAL-logged.
>>>
>>> Perhaps we can just delete that para.  If not, what should it say now?
>>
>> That's obsolete, and can just be removed.
>>
>
> Agreed and posted a patch on hackers [1]

Also the obsolete warning message appears in .po files and the mb tests:

$ git grep 'and their use is discouraged'
src/backend/po/de.po:msgid "hash indexes are not WAL-logged and their
use is discouraged"
src/backend/po/es.po:msgid "hash indexes are not WAL-logged and their
use is discouraged"
src/backend/po/fr.po:msgid "hash indexes are not WAL-logged and their
use is discouraged"
src/backend/po/it.po:msgid "hash indexes are not WAL-logged and their
use is discouraged"
src/backend/po/pl.po:msgid "hash indexes are not WAL-logged and their
use is discouraged"
src/backend/po/ru.po:msgid "hash indexes are not WAL-logged and their
use is discouraged"
src/backend/po/zh_CN.po:msgid "hash indexes are not WAL-logged and
their use is discouraged"
src/test/mb/expected/big5.out:WARNING:  hash indexes are not
WAL-logged and their use is discouraged
src/test/mb/expected/euc_jp.out:WARNING:  hash indexes are not
WAL-logged and their use is discouraged
src/test/mb/expected/euc_kr.out:WARNING:  hash indexes are not
WAL-logged and their use is discouraged
src/test/mb/expected/euc_tw.out:WARNING:  hash indexes are not
WAL-logged and their use is discouraged
src/test/mb/expected/gb18030.out:WARNING:  hash indexes are not
WAL-logged and their use is discouraged
src/test/mb/expected/mule_internal.out:WARNING:  hash indexes are not
WAL-logged and their use is discouraged
src/test/mb/expected/mule_internal.out:WARNING:  hash indexes are not
WAL-logged and their use is discouraged
src/test/mb/expected/sjis.out:WARNING:  hash indexes are not
WAL-logged and their use is discouraged
src/test/mb/expected/utf8.out:WARNING:  hash indexes are not
WAL-logged and their use is discouraged

-- 
Thomas Munro
http://www.enterprisedb.com


-- 
Sent via pgsql-committers mailing list (pgsql-committers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-committers



^ permalink  raw  reply  [nested|flat] 13+ messages in thread

* Re: pgsql: hash: Add write-ahead logging support.
@ 2017-03-15 03:26  Tom Lane <tgl@sss.pgh.pa.us>
  parent: Thomas Munro <thomas.munro@enterprisedb.com>
  0 siblings, 1 reply; 13+ messages in thread

From: Tom Lane @ 2017-03-15 03:26 UTC (permalink / raw)
  To: Thomas Munro <thomas.munro@enterprisedb.com>; +Cc: Amit Kapila <amit.kapila16@gmail.com>; Robert Haas <robertmhaas@gmail.com>; pgsql-committers

Thomas Munro <thomas.munro@enterprisedb.com> writes:
> Also the obsolete warning message appears in .po files and the mb tests:

The .po files won't get updated until the next import from the translation
repo.  No need to worry about them.

However, we probably do need to fix the src/test/mb/expected/ files.
Has anyone tried running that test suite?

			regards, tom lane

-- 
Sent via pgsql-committers mailing list (pgsql-committers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-committers

^ permalink  raw  reply  [nested|flat] 13+ messages in thread

* Re: pgsql: hash: Add write-ahead logging support.
@ 2017-03-15 03:42  Amit Kapila <amit.kapila16@gmail.com>
  parent: Tom Lane <tgl@sss.pgh.pa.us>
  0 siblings, 1 reply; 13+ messages in thread

From: Amit Kapila @ 2017-03-15 03:42 UTC (permalink / raw)
  To: Tom Lane <tgl@sss.pgh.pa.us>; +Cc: Thomas Munro <thomas.munro@enterprisedb.com>; Robert Haas <robertmhaas@gmail.com>; pgsql-committers

On Wed, Mar 15, 2017 at 8:56 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Thomas Munro <thomas.munro@enterprisedb.com> writes:
>> Also the obsolete warning message appears in .po files and the mb tests:
>
> The .po files won't get updated until the next import from the translation
> repo.  No need to worry about them.
>
> However, we probably do need to fix the src/test/mb/expected/ files.
> Has anyone tried running that test suite?
>

I haven't tried.  Does this run as part of any regression tests or is
the only way is run it manually (sh mbregress.sh)?


-- 
With Regards,
Amit Kapila.
EnterpriseDB: http://www.enterprisedb.com


-- 
Sent via pgsql-committers mailing list (pgsql-committers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-committers



^ permalink  raw  reply  [nested|flat] 13+ messages in thread

* Re: pgsql: hash: Add write-ahead logging support.
@ 2017-03-15 03:48  Tom Lane <tgl@sss.pgh.pa.us>
  parent: Amit Kapila <amit.kapila16@gmail.com>
  0 siblings, 1 reply; 13+ messages in thread

From: Tom Lane @ 2017-03-15 03:48 UTC (permalink / raw)
  To: Amit Kapila <amit.kapila16@gmail.com>; +Cc: Thomas Munro <thomas.munro@enterprisedb.com>; Robert Haas <robertmhaas@gmail.com>; pgsql-committers

Amit Kapila <amit.kapila16@gmail.com> writes:
> On Wed, Mar 15, 2017 at 8:56 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> However, we probably do need to fix the src/test/mb/expected/ files.
>> Has anyone tried running that test suite?

> I haven't tried.  Does this run as part of any regression tests or is
> the only way is run it manually (sh mbregress.sh)?

I don't believe it's connected up to any top-level test target ---
if it were, the buildfarm would likely have noticed this.

			regards, tom lane


-- 
Sent via pgsql-committers mailing list (pgsql-committers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-committers



^ permalink  raw  reply  [nested|flat] 13+ messages in thread

* Re: pgsql: hash: Add write-ahead logging support.
@ 2017-03-15 04:14  Amit Kapila <amit.kapila16@gmail.com>
  parent: Tom Lane <tgl@sss.pgh.pa.us>
  0 siblings, 1 reply; 13+ messages in thread

From: Amit Kapila @ 2017-03-15 04:14 UTC (permalink / raw)
  To: Tom Lane <tgl@sss.pgh.pa.us>; +Cc: Thomas Munro <thomas.munro@enterprisedb.com>; Robert Haas <robertmhaas@gmail.com>; pgsql-committers

On Wed, Mar 15, 2017 at 9:18 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Amit Kapila <amit.kapila16@gmail.com> writes:
>> On Wed, Mar 15, 2017 at 8:56 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>>> However, we probably do need to fix the src/test/mb/expected/ files.
>>> Has anyone tried running that test suite?
>
>> I haven't tried.  Does this run as part of any regression tests or is
>> the only way is run it manually (sh mbregress.sh)?
>
> I don't believe it's connected up to any top-level test target ---
> if it were, the buildfarm would likely have noticed this.
>

I also think so.  In any case, I have run those now and they are
failing as expected.  I will send a patch to fix these tests.

-- 
With Regards,
Amit Kapila.
EnterpriseDB: http://www.enterprisedb.com


-- 
Sent via pgsql-committers mailing list (pgsql-committers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-committers



^ permalink  raw  reply  [nested|flat] 13+ messages in thread

* Re: pgsql: hash: Add write-ahead logging support.
@ 2017-03-15 05:49  Amit Kapila <amit.kapila16@gmail.com>
  parent: Amit Kapila <amit.kapila16@gmail.com>
  0 siblings, 0 replies; 13+ messages in thread

From: Amit Kapila @ 2017-03-15 05:49 UTC (permalink / raw)
  To: Tom Lane <tgl@sss.pgh.pa.us>; +Cc: Thomas Munro <thomas.munro@enterprisedb.com>; Robert Haas <robertmhaas@gmail.com>; pgsql-committers

On Wed, Mar 15, 2017 at 9:44 AM, Amit Kapila <amit.kapila16@gmail.com> wrote:
> On Wed, Mar 15, 2017 at 9:18 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> Amit Kapila <amit.kapila16@gmail.com> writes:
>>> On Wed, Mar 15, 2017 at 8:56 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>>>> However, we probably do need to fix the src/test/mb/expected/ files.
>>>> Has anyone tried running that test suite?
>>
>>> I haven't tried.  Does this run as part of any regression tests or is
>>> the only way is run it manually (sh mbregress.sh)?
>>
>> I don't believe it's connected up to any top-level test target ---
>> if it were, the buildfarm would likely have noticed this.
>>
>
> I also think so.  In any case, I have run those now and they are
> failing as expected.  I will send a patch to fix these tests.
>

Attached are two patches, one to fix multi-byte regression tests and
another is spurious function reference in one of the comments in hash
index code.


-- 
With Regards,
Amit Kapila.
EnterpriseDB: http://www.enterprisedb.com

-- 
Sent via pgsql-committers mailing list (pgsql-committers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-committers

Attachments:

fix_mb_test_output_v1.patch (application/octet-stream, 6.3K) download
diff --git a/src/test/mb/expected/big5.out b/src/test/mb/expected/big5.out
index d7a6097..97100cd 100644
--- a/src/test/mb/expected/big5.out
+++ b/src/test/mb/expected/big5.out
@@ -2,7 +2,6 @@ drop table 
 create table ¼t°Ó¸ê®Æ (¦æ·~§O text, ¤½¥q©ïÀY varchar, ¦a§} varchar(16));
 create index ¼t°Ó¸ê®Æindex1 on ¼t°Ó¸ê®Æ using btree (¦æ·~§O);
 create index ¼t°Ó¸ê®Æindex2 on ¼t°Ó¸ê®Æ using hash (¤½¥q©ïÀY);
-WARNING:  hash indexes are not WAL-logged and their use is discouraged
 insert into ¼t°Ó¸ê®Æ values ('¹q¸£·~', '¹F¹F¬ì§Þ', '¥_A01¤¯');
 insert into ¼t°Ó¸ê®Æ values ('»s³y·~', '°]·½¦³¤½¥q', '¤¤B10¤¤');
 insert into ¼t°Ó¸ê®Æ values ('À\¶¼·~', '¬ü¨ýªÑ¥÷¦³¤½¥q', '°ªZ01¤E');
diff --git a/src/test/mb/expected/euc_jp.out b/src/test/mb/expected/euc_jp.out
index be248d9..15f83e0 100644
--- a/src/test/mb/expected/euc_jp.out
+++ b/src/test/mb/expected/euc_jp.out
@@ -3,7 +3,6 @@ ERROR:  table "
 create table ·×»»µ¡ÍÑ¸ì (ÍÑ¸ì text, Ê¬Îà¥³¡¼¥É varchar, È÷¹Í1A¤À¤è char(16));
 create index ·×»»µ¡ÍÑ¸ìindex1 on ·×»»µ¡ÍÑ¸ì using btree (ÍÑ¸ì);
 create index ·×»»µ¡ÍÑ¸ìindex2 on ·×»»µ¡ÍÑ¸ì using hash (Ê¬Îà¥³¡¼¥É);
-WARNING:  hash indexes are not WAL-logged and their use is discouraged
 insert into ·×»»µ¡ÍÑ¸ì values('¥³¥ó¥Ô¥å¡¼¥¿¥Ç¥£¥¹¥×¥ì¥¤','µ¡A01¾å');
 insert into ·×»»µ¡ÍÑ¸ì values('¥³¥ó¥Ô¥å¡¼¥¿¥°¥é¥Õ¥£¥Ã¥¯¥¹','Ê¬B10Ãæ');
 insert into ·×»»µ¡ÍÑ¸ì values('¥³¥ó¥Ô¥å¡¼¥¿¥×¥í¥°¥é¥Þ¡¼','¿ÍZ01²¼');
diff --git a/src/test/mb/expected/euc_kr.out b/src/test/mb/expected/euc_kr.out
index 837f2e6..229ffd8 100644
--- a/src/test/mb/expected/euc_kr.out
+++ b/src/test/mb/expected/euc_kr.out
@@ -3,7 +3,6 @@ ERROR:  table "Íªß©Ñ¦
 create table Íªß©Ñ¦¿ë¾î (¿ë¾î text, ÝÂ×¾ÄÚµå varchar, ºñ°í1A¶ó±¸ char(16));
 create index Íªß©Ñ¦¿ë¾îindex1 on Íªß©Ñ¦¿ë¾î using btree (¿ë¾î);
 create index Íªß©Ñ¦¿ë¾îindex2 on Íªß©Ñ¦¿ë¾î using hash (ÝÂ×¾ÄÚµå);
-WARNING:  hash indexes are not WAL-logged and their use is discouraged
 insert into Íªß©Ñ¦¿ë¾î values('ÄÄÇ»ÅÍµð½ºÇÃ·¹ÀÌ', 'Ñ¦A01ß¾');
 insert into Íªß©Ñ¦¿ë¾î values('ÄÄÇ»ÅÍ±×·¡ÇÈ½º', 'ÝÂB10ñé');
 insert into Íªß©Ñ¦¿ë¾î values('ÄÄÇ»ÅÍÇÁ·Î±×·¡¸Ó', 'ìÑZ01ù»');
diff --git a/src/test/mb/expected/euc_tw.out b/src/test/mb/expected/euc_tw.out
index f224e78..99eb488 100644
--- a/src/test/mb/expected/euc_tw.out
+++ b/src/test/mb/expected/euc_tw.out
@@ -3,7 +3,6 @@ ERROR:  table "
 create table ìÞÙ¸æñÕè (ÈçäÆÉ± text, ÄüÆ³Íñó¤ varchar, ÇâÉß varchar(16));
 create index ìÞÙ¸æñÕèindex1 on ìÞÙ¸æñÕè using btree (ÈçäÆÉ±);
 create index ìÞÙ¸æñÕèindex2 on ìÞÙ¸æñÕè using hash (ÄüÆ³Íñó¤);
-WARNING:  hash indexes are not WAL-logged and their use is discouraged
 insert into ìÞÙ¸æñÕè values ('çÙæªäÆ', 'ç®ç®ÒïÊÀ', 'Æ¡A01Äî');
 insert into ìÞÙ¸æñÕè values ('êûÝßäÆ', 'ØÂäãÈ´ÓîÄüÆ³', 'ÄãB10Äã');
 insert into ìÞÙ¸æñÕè values ('ó§ã£äÆ', 'Ó¡ÌÀÏ´Ç¹È´ÓîÄüÆ³', 'ØíZ01Ä¦');
diff --git a/src/test/mb/expected/gb18030.out b/src/test/mb/expected/gb18030.out
index 562852b..b14234d 100644
--- a/src/test/mb/expected/gb18030.out
+++ b/src/test/mb/expected/gb18030.out
@@ -2,7 +2,6 @@ drop table Ó
 create table ÓËãCÓÃÕZ (ÓÃÕZ text, ·Öî¥³©`¥É varchar, ä¿¼1A¤À¤è char(16));
 create index ÓËãCÓÃÕZindex1 on ÓËãCÓÃÕZ using btree (ÓÃÕZ);
 create index ÓËãCÓÃÕZindex2 on ÓËãCÓÃÕZ using hash (·Öî¥³©`¥É);
-WARNING:  hash indexes are not WAL-logged and their use is discouraged
 insert into ÓËãCÓÃÕZ values('¥³¥ó¥Ô¥å©`¥¿¥Ç¥£¥¹¥×¥ì¥¤','CA01ÉÏ');
 insert into ÓËãCÓÃÕZ values('¥³¥ó¥Ô¥å©`¥¿¥°¥é¥Õ¥£¥Ã¥¯¥¹','·ÖB10ÖÐ');
 insert into ÓËãCÓÃÕZ values('¥³¥ó¥Ô¥å©`¥¿¥×¥í¥°¥é¥Þ©`','ÈËZ01ÏÂ');
diff --git a/src/test/mb/expected/mule_internal.out b/src/test/mb/expected/mule_internal.out
index 67f7775..ac8b57d 100644
--- a/src/test/mb/expected/mule_internal.out
+++ b/src/test/mb/expected/mule_internal.out
@@ -3,7 +3,6 @@ ERROR:  table "
 create table ·×»»µ¡ÍÑ¸ì (ÍÑ¸ì text, Ê¬Îà¥³¡¼¥É varchar, È÷¹Í1A¤À¤è char(16));
 create index ·×»»µ¡ÍÑ¸ìindex1 on ·×»»µ¡ÍÑ¸ì using btree (ÍÑ¸ì);
 create index ·×»»µ¡ÍÑ¸ìindex2 on ·×»»µ¡ÍÑ¸ì using hash (Ê¬Îà¥³¡¼¥É);
-WARNING:  hash indexes are not WAL-logged and their use is discouraged
 insert into ·×»»µ¡ÍÑ¸ì values('¥³¥ó¥Ô¥å¡¼¥¿¥Ç¥£¥¹¥×¥ì¥¤','µ¡A01¾å');
 insert into ·×»»µ¡ÍÑ¸ì values('¥³¥ó¥Ô¥å¡¼¥¿¥°¥é¥Õ¥£¥Ã¥¯¥¹','Ê¬B10Ãæ');
 insert into ·×»»µ¡ÍÑ¸ì values('¥³¥ó¥Ô¥å¡¼¥¿¥×¥í¥°¥é¥Þ¡¼','¿ÍZ01²¼');
@@ -178,7 +177,6 @@ ERROR:  table "
 create table Íªß©Ñ¦¿ë¾î (¿ë¾î text, ÝÂ×¾ÄÚµå varchar, ºñ°í1A¶ó±¸ char(16));
 create index Íªß©Ñ¦¿ë¾îindex1 on Íªß©Ñ¦¿ë¾î using btree (¿ë¾î);
 create index Íªß©Ñ¦¿ë¾îindex2 on Íªß©Ñ¦¿ë¾î using hash (ÝÂ×¾ÄÚµå);
-WARNING:  hash indexes are not WAL-logged and their use is discouraged
 insert into Íªß©Ñ¦¿ë¾î values('ÄÄÇ»ÅÍµð½ºÇÃ·¹ÀÌ', 'Ñ¦A01ß¾');
 insert into Íªß©Ñ¦¿ë¾î values('ÄÄÇ»ÅÍ±×·¡ÇÈ½º', 'ÝÂB10ñé');
 insert into Íªß©Ñ¦¿ë¾î values('ÄÄÇ»ÅÍÇÁ·Î±×·¡¸Ó', 'ìÑZ01ù»');
diff --git a/src/test/mb/expected/sjis.out b/src/test/mb/expected/sjis.out
index 73dd29b..db2bfd0 100644
--- a/src/test/mb/expected/sjis.out
+++ b/src/test/mb/expected/sjis.out
@@ -2,7 +2,6 @@ drop table 
 create table vZ@pê (pê text, ªÞR[h varchar, õl1A¾æ char(16));
 create index vZ@pêindex1 on vZ@pê using btree (pê);
 create index vZ@pêindex2 on vZ@pê using hash (ªÞR[h);
-WARNING:  hash indexes are not WAL-logged and their use is discouraged
 insert into vZ@pê values('Rs[^fBXvC','@A01ã');
 insert into vZ@pê values('Rs[^OtBbNX','ªB10');
 insert into vZ@pê values('Rs[^vO}[','lZ01º');
diff --git a/src/test/mb/expected/utf8.out b/src/test/mb/expected/utf8.out
index 715f346..8f9f63c 100644
--- a/src/test/mb/expected/utf8.out
+++ b/src/test/mb/expected/utf8.out
@@ -3,7 +3,6 @@ ERROR:  table "è¨ç®æ©ç¨èª" does not exist
 create table è¨ç®æ©ç¨èª (ç¨èª text, åé¡ã³ã¼ã varchar, åè1Aã ã char(16));
 create index è¨ç®æ©ç¨èªindex1 on è¨ç®æ©ç¨èª using btree (ç¨èª);
 create index è¨ç®æ©ç¨èªindex2 on è¨ç®æ©ç¨èª using hash (åé¡ã³ã¼ã);
-WARNING:  hash indexes are not WAL-logged and their use is discouraged
 insert into è¨ç®æ©ç¨èª values('ã³ã³ãã¥ã¼ã¿ãã£ã¹ãã¬ã¤','æ©A01ä¸');
 insert into è¨ç®æ©ç¨èª values('ã³ã³ãã¥ã¼ã¿ã°ã©ãã£ãã¯ã¹','åB10ä¸');
 insert into è¨ç®æ©ç¨èª values('ã³ã³ãã¥ã¼ã¿ããã°ã©ãã¼','äººZ01ä¸');


change_spurious_function_v1.patch (application/octet-stream, 732B) download
diff --git a/src/backend/access/hash/hashpage.c b/src/backend/access/hash/hashpage.c
index dc606f1..622cc4b 100644
--- a/src/backend/access/hash/hashpage.c
+++ b/src/backend/access/hash/hashpage.c
@@ -975,7 +975,7 @@ fail:
  * hash indexes sequentially anyway, that probably doesn't matter.
  *
  * XXX It's annoying that this code is executed with the metapage lock held.
- * We need to interlock against _hash_getovflpage() adding a new overflow page
+ * We need to interlock against _hash_addovflpage() adding a new overflow page
  * concurrently, but it'd likely be better to use LockRelationForExtension
  * for the purpose.  OTOH, adding a splitpoint is a very infrequent operation,
  * so it may not be worth worrying about.


^ permalink  raw  reply  [nested|flat] 13+ messages in thread

end of thread, other threads:[~2017-03-15 05:49 UTC | newest]

Thread overview: 13+ messages (download: mbox mbox.gz follow: Atom feed)
-- links below jump to the message on this page --
2017-03-14 17:27 pgsql: hash: Add write-ahead logging support. Robert Haas <rhaas@postgresql.org>
2017-03-14 17:54 ` Tom Lane <tgl@sss.pgh.pa.us>
2017-03-14 17:56   ` Robert Haas <robertmhaas@gmail.com>
2017-03-14 18:18     ` Tom Lane <tgl@sss.pgh.pa.us>
2017-03-14 18:22       ` Robert Haas <robertmhaas@gmail.com>
2017-03-14 18:39         ` Tom Lane <tgl@sss.pgh.pa.us>
2017-03-15 02:23     ` Amit Kapila <amit.kapila16@gmail.com>
2017-03-15 02:35       ` Thomas Munro <thomas.munro@enterprisedb.com>
2017-03-15 03:26         ` Tom Lane <tgl@sss.pgh.pa.us>
2017-03-15 03:42           ` Amit Kapila <amit.kapila16@gmail.com>
2017-03-15 03:48             ` Tom Lane <tgl@sss.pgh.pa.us>
2017-03-15 04:14               ` Amit Kapila <amit.kapila16@gmail.com>
2017-03-15 05:49                 ` Amit Kapila <amit.kapila16@gmail.com>

This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox