[COMMITTERS] pgsql: Count locked pages that don't need vacuuming as scanned.

2013-11-18 Thread Heikki Linnakangas
Count locked pages that don't need vacuuming as scanned.

Previously, if VACUUM skipped vacuuming a page because it's pinned, it
didn't count that page as scanned. However, that meant that relfrozenxid
was not bumped up either, which prevented anti-wraparound vacuum from
doing its job.

Report by Миша Тюрин, analysis and patch by Sergey Burladyn and Jeff Janes.
Backpatch to 9.2, where the skip-locked-pages behavior was introduced.

Branch
--
REL9_3_STABLE

Details
---
http://git.postgresql.org/pg/commitdiff/ea2bb1b47d7629a17dbc0c7da66cf063f8d3a768

Modified Files
--
src/backend/commands/vacuumlazy.c |1 +
1 file changed, 1 insertion(+)


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


[COMMITTERS] pgsql: Count locked pages that don't need vacuuming as scanned.

2013-11-18 Thread Heikki Linnakangas
Count locked pages that don't need vacuuming as scanned.

Previously, if VACUUM skipped vacuuming a page because it's pinned, it
didn't count that page as scanned. However, that meant that relfrozenxid
was not bumped up either, which prevented anti-wraparound vacuum from
doing its job.

Report by Миша Тюрин, analysis and patch by Sergey Burladyn and Jeff Janes.
Backpatch to 9.2, where the skip-locked-pages behavior was introduced.

Branch
--
REL9_2_STABLE

Details
---
http://git.postgresql.org/pg/commitdiff/3379263b6ded774acb96288d1e67caa37dcba0de

Modified Files
--
src/backend/commands/vacuumlazy.c |1 +
1 file changed, 1 insertion(+)


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


[COMMITTERS] pgsql: Count locked pages that don't need vacuuming as scanned.

2013-11-18 Thread Heikki Linnakangas
Count locked pages that don't need vacuuming as scanned.

Previously, if VACUUM skipped vacuuming a page because it's pinned, it
didn't count that page as scanned. However, that meant that relfrozenxid
was not bumped up either, which prevented anti-wraparound vacuum from
doing its job.

Report by Миша Тюрин, analysis and patch by Sergey Burladyn and Jeff Janes.
Backpatch to 9.2, where the skip-locked-pages behavior was introduced.

Branch
--
master

Details
---
http://git.postgresql.org/pg/commitdiff/4c697d8f4845823a8af67788b219ffa4516ad14c

Modified Files
--
src/backend/commands/vacuumlazy.c |1 +
1 file changed, 1 insertion(+)


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


[COMMITTERS] pgsql: Use cstring_to_text_with_len when length is known.

2013-11-18 Thread Robert Haas
Use cstring_to_text_with_len when length is known.

This avoids a potentially-expensive extra call to strlen().

David Rowley

Branch
--
master

Details
---
http://git.postgresql.org/pg/commitdiff/f1df4731eea6bc05e0769e9cc789e7304722efe4

Modified Files
--
contrib/pageinspect/fsmfuncs.c |2 +-
src/backend/utils/adt/json.c   |   12 ++--
2 files changed, 7 insertions(+), 7 deletions(-)


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


[COMMITTERS] pgsql: Replace appendPQExpBuffer(..., ) with appendPQExpBuffe

2013-11-18 Thread Heikki Linnakangas
Replace appendPQExpBuffer(..., ) with appendPQExpBufferStr

Arguably makes the code a bit more readable, and might give a small
performance gain.

David Rowley

Branch
--
master

Details
---
http://git.postgresql.org/pg/commitdiff/32ceba3ea730b6b1bd3eca786f72d61945ad42b7

Modified Files
--
src/bin/pg_dump/dumputils.c  |   18 +-
src/bin/pg_dump/pg_backup_archiver.c |   12 +-
src/bin/pg_dump/pg_dump.c|  725 +-
src/bin/pg_dump/pg_dumpall.c |  116 +++---
src/bin/psql/command.c   |2 +-
src/bin/psql/copy.c  |4 +-
src/bin/psql/describe.c  |  520 
src/bin/psql/tab-complete.c  |4 +-
src/bin/scripts/clusterdb.c  |6 +-
src/bin/scripts/createdb.c   |4 +-
src/bin/scripts/createuser.c |   32 +-
src/bin/scripts/reindexdb.c  |4 +-
src/bin/scripts/vacuumdb.c   |   18 +-
src/interfaces/libpq/fe-connect.c|   22 +-
src/test/isolation/isolationtester.c |2 +-
15 files changed, 739 insertions(+), 750 deletions(-)


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


[COMMITTERS] pgsql: Spell SQL keywords in uppercase in pg_dump's query.

2013-11-18 Thread Heikki Linnakangas
Spell SQL keywords in uppercase in pg_dump's query.

The server won't care, but let's be consistent.

David Rowley.

Branch
--
master

Details
---
http://git.postgresql.org/pg/commitdiff/fea437681d68a0027091c2d534c69c0965cf06f1

Modified Files
--
src/bin/pg_dump/pg_dump.c |   48 ++---
1 file changed, 24 insertions(+), 24 deletions(-)


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


[COMMITTERS] pgsql: unaccent: Revert patch 9299f6179838cef8aa1123f6fb76f0d3d6f2decc

2013-11-18 Thread Bruce Momjian
unaccent:  Revert patch 9299f6179838cef8aa1123f6fb76f0d3d6f2decc

The reverted patch to change functions from strict to immutable was
incorrect and needs additional research.

Branch
--
master

Details
---
http://git.postgresql.org/pg/commitdiff/0dbf9a6a91675b99a5097528729c38e7f3015646

Modified Files
--
contrib/unaccent/unaccent--1.0.sql |4 ++--
contrib/unaccent/unaccent--unpackaged--1.0.sql |6 +++---
2 files changed, 5 insertions(+), 5 deletions(-)


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


Re: [COMMITTERS] pgsql: Ensure that XLOG_HEAP2_VISIBLE always targets an initialized pag

2013-11-18 Thread Bruce Momjian
On Sun, Nov 10, 2013 at 11:51:06PM +0100, Andres Freund wrote:
> Hi,
> 
> On 2013-06-06 14:22:12 +, Robert Haas wrote:
> > Ensure that XLOG_HEAP2_VISIBLE always targets an initialized page.
> 
> This didn't make it to the release notes for 9.2.5. Since there have
> been several people asking for it since, can we maybe add it
> retroactively?

OK, what text would I use?  This?

* Prevent unitialized heap pages from causing standby replay
  failures

-- 
  Bruce Momjian  http://momjian.us
  EnterpriseDB http://enterprisedb.com

  + Everyone has their own god. +


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


Re: [COMMITTERS] pgsql: Ensure that XLOG_HEAP2_VISIBLE always targets an initialized pag

2013-11-18 Thread Andres Freund
On 2013-11-18 16:33:25 -0500, Bruce Momjian wrote:
> On Sun, Nov 10, 2013 at 11:51:06PM +0100, Andres Freund wrote:
> > Hi,
> > 
> > On 2013-06-06 14:22:12 +, Robert Haas wrote:
> > > Ensure that XLOG_HEAP2_VISIBLE always targets an initialized page.
> > 
> > This didn't make it to the release notes for 9.2.5. Since there have
> > been several people asking for it since, can we maybe add it
> > retroactively?
> 
> OK, what text would I use?  This?
> 
> * Prevent unitialized heap pages from causing standby replay
>   failures

Maybe:

* Prevent errors in WAL replay due to references to unitilizated empty
  pages in vacuum.

Greetings,

Andres Freund

-- 
 Andres Freund http://www.2ndQuadrant.com/
 PostgreSQL Development, 24x7 Support, Training & Services


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


[COMMITTERS] pgsql: pg_upgrade: Report full disk better

2013-11-18 Thread Peter Eisentraut
pg_upgrade: Report full disk better

Previously, pg_upgrade would abort copy_file() on a short write without
setting errno, which the caller would report as an error with the
message "Success".  We assume ENOSPC in that case, as we do elsewhere in
the code.  Also set errno in some other error cases in copy_file() to
avoid bogus "Success" error messages.

This was broken in 6b711cf37c228749b6a8cef50e16e3c587d18dd4, so 9.2 and
before are OK.

Branch
--
master

Details
---
http://git.postgresql.org/pg/commitdiff/bd5a9a50e4946ffc58a632b5717c8cfddd3ea6ce

Modified Files
--
contrib/pg_upgrade/file.c |9 +
1 file changed, 9 insertions(+)


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


[COMMITTERS] pgsql: pg_upgrade: Report full disk better

2013-11-18 Thread Peter Eisentraut
pg_upgrade: Report full disk better

Previously, pg_upgrade would abort copy_file() on a short write without
setting errno, which the caller would report as an error with the
message "Success".  We assume ENOSPC in that case, as we do elsewhere in
the code.  Also set errno in some other error cases in copy_file() to
avoid bogus "Success" error messages.

This was broken in 6b711cf37c228749b6a8cef50e16e3c587d18dd4, so 9.2 and
before are OK.

Branch
--
REL9_3_STABLE

Details
---
http://git.postgresql.org/pg/commitdiff/8ca75671bd583f2e814c865941da464d2419312c

Modified Files
--
contrib/pg_upgrade/file.c |9 +
1 file changed, 9 insertions(+)


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