Re: [HACKERS] discarding duplicate indexes

2012-12-20 Thread Gavin Flower
On 20/12/12 14:57, Josh Kupershmidt wrote: CREATE TABLE test (id int); CREATE INDEX test_idx1 ON test (id); CREATE INDEX test_idx2 ON test (id); I initially misread your example code, but after I realised my mistake, I thought of an alternative scenario that might be worth considering. CREATE

Re: [HACKERS] [GENERAL] trouble with pg_upgrade 9.0 - 9.1

2012-12-20 Thread Bruce Momjian
19.12.2012, 21:47, Tom Lane t...@sss.pgh.pa.us: Kevin Grittner kgri...@mail.com writes:  Groshev Andrey wrote:    Mismatch of relation names: database database, old rel public.lob.ВерсияВнешнегоДокумента$Документ_pkey, new rel public.plob.ВерсияВнешнегоДокумента$Документ  There is

Re: [HACKERS] discarding duplicate indexes

2012-12-20 Thread John R Pierce
On 12/20/2012 12:26 AM, Gavin Flower wrote: CREATE TABLE test (id int, int sub, text payload); CREATE INDEX test_idx1 ON test (id, sub); CREATE INDEX test_idx2 ON test (id); Nowtest_idx2 is logically included intest_idx1, but if the majority of transactions only query onid, thentest_idx2

Re: [HACKERS] Review of Row Level Security

2012-12-20 Thread Simon Riggs
On 20 December 2012 00:24, Robert Haas robertmh...@gmail.com wrote: On Wed, Dec 19, 2012 at 12:54 PM, Simon Riggs si...@2ndquadrant.com wrote: I can see a use case for not having security apply for users who have *only* INSERT privilege. This would allow people to run bulk loads of data into a

Re: [HACKERS] Set visibility map bit after HOT prune

2012-12-20 Thread Simon Riggs
On 20 December 2012 00:43, Robert Haas robertmh...@gmail.com wrote: On Wed, Dec 19, 2012 at 12:39 PM, Simon Riggs si...@2ndquadrant.com wrote: The benefit of saying that only UPDATEs clean the block is that this penalises only the workload making the mess, rather than everybody cleaning up

Re: [HACKERS] [GENERAL] trouble with pg_upgrade 9.0 - 9.1

2012-12-20 Thread Bruce Momjian
On Thu, Dec 20, 2012 at 08:55:16AM +0400, Groshev Andrey wrote: No, old database not use table plob.. only primary key -- -- Name: plob.ВерсияВнешнегоДокумента$Документ; Type: CONSTRAINT; Schema: public; Owner: postgres; Tablespace: -- -- For binary upgrade, must preserve

Re: [HACKERS] Set visibility map bit after HOT prune

2012-12-20 Thread Simon Riggs
On 20 December 2012 00:10, Pavan Deolasee pavan.deola...@gmail.com wrote: I just thought that we can fairly easily limit the damage if we are really worried about SELECTs being penalised. What if we set a configurable limit on *extra* things that a query may do which is otherwise not very

[HACKERS] EDB hosted buildfarm animals - extended downtime

2012-12-20 Thread Dave Page
Due to circumstances beyond my control (blame the power company), the following buildfarm animals will be down from 27th December until 2nd January: baiji mastodon protosciurus castoroides -- Dave Page Blog: http://pgsnake.blogspot.com Twitter: @pgsnake EnterpriseDB UK:

Re: [HACKERS] ThisTimeLineID in checkpointer and bgwriter processes

2012-12-20 Thread Amit Kapila
On Wednesday, December 19, 2012 9:30 PM Heikki Linnakangas wrote: In both checkpointer.c and bgwriter.c, we do this before entering the main loop: /* * Use the recovery target timeline ID during recovery */ if (RecoveryInProgress())

Re: [HACKERS] [GENERAL] trouble with pg_upgrade 9.0 - 9.1

2012-12-20 Thread Groshev Andrey
20.12.2012, 13:00, Bruce Momjian br...@momjian.us: On Thu, Dec 20, 2012 at 08:55:16AM +0400, Groshev Andrey wrote:  No, old database not use table plob..  only primary key  --  -- Name: plob.ВерсияВнешнегоДокумента$Документ; Type: CONSTRAINT; Schema: public; Owner: postgres;

Re: [HACKERS] PATCH: optimized DROP of multiple tables within a transaction

2012-12-20 Thread Andres Freund
On 2012-12-20 02:54:48 +0100, Tomas Vondra wrote: On 19.12.2012 02:18, Andres Freund wrote: On 2012-12-17 00:31:00 +0100, Tomas Vondra wrote: I think except of the temp buffer issue mentioned below its ready. -DropRelFileNodeAllBuffers(RelFileNodeBackend rnode)

Re: [HACKERS] [GENERAL] trouble with pg_upgrade 9.0 - 9.1

2012-12-20 Thread Bruce Momjian
On Thu, Dec 20, 2012 at 03:19:17PM +0400, Groshev Andrey wrote: 20.12.2012, 13:00, Bruce Momjian br...@momjian.us: On Thu, Dec 20, 2012 at 08:55:16AM +0400, Groshev Andrey wrote:  No, old database not use table plob..  only primary key  --  -- Name:

Re: [HACKERS] ThisTimeLineID in checkpointer and bgwriter processes

2012-12-20 Thread Heikki Linnakangas
On 20.12.2012 12:08, Amit Kapila wrote: On Wednesday, December 19, 2012 9:30 PM Heikki Linnakangas wrote: In both checkpointer.c and bgwriter.c, we do this before entering the main loop: /* * Use the recovery target timeline ID during recovery */ if

Re: [HACKERS] [GENERAL] trouble with pg_upgrade 9.0 - 9.1

2012-12-20 Thread Groshev Andrey
20.12.2012, 11:43, Bruce Momjian br...@momjian.us:  19.12.2012, 21:47, Tom Lane t...@sss.pgh.pa.us:  Kevin Grittner kgri...@mail.com writes:   Groshev Andrey wrote:     Mismatch of relation names: database database, old rel public.lob.ВерсияВнешнегоДокумента$Документ_pkey, new rel

Re: [HACKERS] [GENERAL] trouble with pg_upgrade 9.0 - 9.1

2012-12-20 Thread Bruce Momjian
On Thu, Dec 20, 2012 at 03:41:37PM +0400, Groshev Andrey wrote: See that 786665369?  That is the pg_class.oid of the plob in the old cluster, and hopefully the new one.  Find where the lob*_pkey index is created and get that oid.  Those should match the same names of the pg_class.oid in

Re: [HACKERS] Re: patch submission: truncate trailing nulls from heap rows to reduce the size of the null bitmap [Review]

2012-12-20 Thread Simon Riggs
On 13 October 2012 08:54, Amit kapila amit.kap...@huawei.com wrote: As per the last discussion for this patch, performance data needs to be provided before this patch's Review can proceed further. So as per your suggestion and from the discussions about this patch, I have collected the

[HACKERS] pg_top

2012-12-20 Thread Brett Maton
Hi List, This might not be the right place to post, but I've got a minor patch for pg_top and would like to submit it for review. Basically, the rpm version in the repositories pg_top92-3.6.2 doesn't work with Postgres 9.2 #define QUERY_PROCESSES \ SELECT

Re: [HACKERS] [GENERAL] trouble with pg_upgrade 9.0 - 9.1

2012-12-20 Thread Bruce Momjian
On Wed, Dec 19, 2012 at 10:19:30PM -0500, Bruce Momjian wrote: On Wed, Dec 19, 2012 at 12:56:05PM -0500, Kevin Grittner wrote: Groshev Andrey wrote:   Mismatch of relation names: database database, old rel public.lob.ВерсияВнешнегоДокумента$Документ_pkey, new rel

Re: [HACKERS] pg_top

2012-12-20 Thread P. Christeas
On Thursday 20 December 2012, Brett Maton wrote: Hi List, This might not be the right place to post, but I've got a minor patch for pg_top and would like to submit it for review. This project lies on Github: https://github.com/markwkm/pg_top/ I think, this commit had fixed the

Re: [HACKERS] Switching timeline over streaming replication

2012-12-20 Thread Heikki Linnakangas
On 17.12.2012 15:05, Thom Brown wrote: I just set up 120 chained standbys, and for some reason I'm seeing these errors: LOG: replication terminated by primary server DETAIL: End of WAL reached on timeline 1 LOG: record with zero length at 0/301EC10 LOG: fetching timeline history file for

Re: [HACKERS] pg_top

2012-12-20 Thread Brett Maton
Ah, I cloned the repository from git://git.postgresql.org/ I'll have a look at Marks repo. Thanks, Brett -Original Message- From: pgsql-hackers-ow...@postgresql.org [mailto:pgsql-hackers-ow...@postgresql.org] On Behalf Of P. Christeas Sent: 20 December 2012 12:33 To:

Re: [HACKERS] Switching timeline over streaming replication

2012-12-20 Thread Andres Freund
On 2012-12-20 14:45:05 +0200, Heikki Linnakangas wrote: On 17.12.2012 15:05, Thom Brown wrote: I just set up 120 chained standbys, and for some reason I'm seeing these errors: LOG: replication terminated by primary server DETAIL: End of WAL reached on timeline 1 LOG: record with zero

Re: [HACKERS] Parser Cruft in gram.y

2012-12-20 Thread Robert Haas
On Wed, Dec 19, 2012 at 10:18 PM, Tom Lane t...@sss.pgh.pa.us wrote: valgrind comes with a tool called cachegrind which can emulate the cache algorithm on some variants of various cpus and produce reports. Can it be made to produce a report for a specific block of memory? I believe that

Re: [HACKERS] Cascading replication: should we detect/prevent cycles?

2012-12-20 Thread Robert Haas
On Wed, Dec 19, 2012 at 5:14 PM, Joshua Berkus j...@agliodbs.com wrote: What would such a test look like? It's not obvious to me that there's any rapid way for a user to detect this situation, without checking each server individually. Change something on the master and observe that none of

Re: [HACKERS] ThisTimeLineID in checkpointer and bgwriter processes

2012-12-20 Thread Amit Kapila
On Thursday, December 20, 2012 5:12 PM Heikki Linnakangas wrote: On 20.12.2012 12:08, Amit Kapila wrote: On Wednesday, December 19, 2012 9:30 PM Heikki Linnakangas wrote: In both checkpointer.c and bgwriter.c, we do this before entering the main loop: /* * Use the

Re: [HACKERS] ThisTimeLineID in checkpointer and bgwriter processes

2012-12-20 Thread Simon Riggs
On 20 December 2012 13:19, Amit Kapila amit.kap...@huawei.com wrote: So I think we're good on that front. But I'll add a comment there, and use 0 explicitly instead of ThisTimeLineID, for clarity. True, it might not have any functionality effect in RemoveOldXlogFiles(). However it can be

Re: [HACKERS] ThisTimeLineID in checkpointer and bgwriter processes

2012-12-20 Thread Andres Freund
On 2012-12-20 13:32:36 +, Simon Riggs wrote: On 20 December 2012 13:19, Amit Kapila amit.kap...@huawei.com wrote: So I think we're good on that front. But I'll add a comment there, and use 0 explicitly instead of ThisTimeLineID, for clarity. True, it might not have any functionality

Re: [HACKERS] Parser Cruft in gram.y

2012-12-20 Thread Peter Eisentraut
On 12/18/12 5:10 PM, Robert Haas wrote: I can't help but suspect that the way we handle keywords today is monumentally inefficient. The unreserved_keyword products, et al, just seem somehow badly wrong-headed. We take the trouble to distinguish all of those cases so that we an turn around

Re: [HACKERS] Parser Cruft in gram.y

2012-12-20 Thread Simon Riggs
On 18 December 2012 22:10, Robert Haas robertmh...@gmail.com wrote: Well that would be nice, but the problem is that I see no way to implement it. If, with a unified parser, the parser is 14% of our source code, then splitting it in two will probably crank that number up well over 20%,

Re: [HACKERS] ThisTimeLineID in checkpointer and bgwriter processes

2012-12-20 Thread Amit Kapila
On Thursday, December 20, 2012 7:03 PM Simon Riggs wrote: On 20 December 2012 13:19, Amit Kapila amit.kap...@huawei.com wrote: So I think we're good on that front. But I'll add a comment there, and use 0 explicitly instead of ThisTimeLineID, for clarity. True, it might not have any

Re: [HACKERS] Parser Cruft in gram.y

2012-12-20 Thread Robert Haas
On Thu, Dec 20, 2012 at 8:55 AM, Simon Riggs si...@2ndquadrant.com wrote: On 18 December 2012 22:10, Robert Haas robertmh...@gmail.com wrote: Well that would be nice, but the problem is that I see no way to implement it. If, with a unified parser, the parser is 14% of our source code, then

[HACKERS] recent ALTER whatever .. SET SCHEMA refactoring

2012-12-20 Thread Robert Haas
The recent SET SCHEMA refactoring has changed the error message that you get when trying to move a function into the schema that already contains it. For a table, as ever, you get: rhaas=# create table foo (a int); CREATE TABLE rhaas=# alter table foo set schema public; ERROR: table foo is

Re: [HACKERS] Parser Cruft in gram.y

2012-12-20 Thread Andres Freund
On 2012-12-20 09:11:46 -0500, Robert Haas wrote: On Thu, Dec 20, 2012 at 8:55 AM, Simon Riggs si...@2ndquadrant.com wrote: On 18 December 2012 22:10, Robert Haas robertmh...@gmail.com wrote: Well that would be nice, but the problem is that I see no way to implement it. If, with a unified

Re: [HACKERS] Parser Cruft in gram.y

2012-12-20 Thread Andres Freund
On 2012-12-20 15:45:47 +0100, Andres Freund wrote: On 2012-12-20 09:11:46 -0500, Robert Haas wrote: On Thu, Dec 20, 2012 at 8:55 AM, Simon Riggs si...@2ndquadrant.com wrote: On 18 December 2012 22:10, Robert Haas robertmh...@gmail.com wrote: Well that would be nice, but the problem is

Re: [HACKERS] Re: patch submission: truncate trailing nulls from heap rows to reduce the size of the null bitmap [Review]

2012-12-20 Thread Amit Kapila
On Thursday, December 20, 2012 5:46 PM Simon Riggs wrote: On 13 October 2012 08:54, Amit kapila amit.kap...@huawei.com wrote: As per the last discussion for this patch, performance data needs to be provided before this patch's Review can proceed further. So as per your suggestion and

Re: [HACKERS] Parser Cruft in gram.y

2012-12-20 Thread Andres Freund
On 2012-12-20 15:51:37 +0100, Andres Freund wrote: On 2012-12-20 15:45:47 +0100, Andres Freund wrote: On 2012-12-20 09:11:46 -0500, Robert Haas wrote: On Thu, Dec 20, 2012 at 8:55 AM, Simon Riggs si...@2ndquadrant.com wrote: On 18 December 2012 22:10, Robert Haas

Re: [HACKERS] discarding duplicate indexes

2012-12-20 Thread Josh Kupershmidt
On Thu, Dec 20, 2012 at 1:26 AM, Gavin Flower gavinflo...@archidevsys.co.nz wrote: On 20/12/12 14:57, Josh Kupershmidt wrote: CREATE TABLE test (id int); CREATE INDEX test_idx1 ON test (id); CREATE INDEX test_idx2 ON test (id); I initially misread your example code, but after I realised my

Re: [HACKERS] Parser Cruft in gram.y

2012-12-20 Thread Andres Freund
On 2012-12-20 16:04:49 +0100, Andres Freund wrote: On 2012-12-20 15:51:37 +0100, Andres Freund wrote: When doing a source/assembly annotation in SearchCatCache about half of the misses are attributed to the memcpy() directly at the beginning. Using a separate array for storing the arguments

Re: [HACKERS] operator dependency of commutator and negator, redux

2012-12-20 Thread Tom Lane
Brendan Jurd dire...@gmail.com writes: On 20 December 2012 11:51, Tom Lane t...@sss.pgh.pa.us wrote: While reconsidering the various not-too-satisfactory fixes we thought of back then, I had a sudden thought. Instead of having a COMMUTATOR or NEGATOR forward reference create a shell operator

Re: [HACKERS] Set visibility map bit after HOT prune

2012-12-20 Thread Robert Haas
On Wed, Dec 19, 2012 at 11:12 PM, Pavan Deolasee pavan.deola...@gmail.com wrote: Yeah, VM buffer contention can become prominent if we break the invariant that page level bit status implies the vm bit status, at least when its clear.OTOH IMHO we need some mechanism to address the issue of

Re: [HACKERS] Set visibility map bit after HOT prune

2012-12-20 Thread Robert Haas
On Thu, Dec 20, 2012 at 4:58 AM, Simon Riggs si...@2ndquadrant.com wrote: ISTM that if someone spots a block that could use cleanup, they mark the block as BM_PIN_COUNT_WAITER, but don't set pid. Then when they unpin the block they send a signal/queue work for a special cleaning process to

Re: [HACKERS] operator dependency of commutator and negator, redux

2012-12-20 Thread Robert Haas
On Thu, Dec 20, 2012 at 10:52 AM, Tom Lane t...@sss.pgh.pa.us wrote: Brendan Jurd dire...@gmail.com writes: On 20 December 2012 11:51, Tom Lane t...@sss.pgh.pa.us wrote: While reconsidering the various not-too-satisfactory fixes we thought of back then, I had a sudden thought. Instead of

Re: [HACKERS] Parser Cruft in gram.y

2012-12-20 Thread Greg Stark
On Thu, Dec 20, 2012 at 3:18 AM, Tom Lane t...@sss.pgh.pa.us wrote: Greg Stark st...@mit.edu writes: But I'm not entirely convinced any of this is actually useful. Just becuase the transition table is large doesn't mean it's inefficient. That's a fair point. However, I've often noticed

Re: [HACKERS] [GENERAL] trouble with pg_upgrade 9.0 - 9.1

2012-12-20 Thread Tom Lane
Bruce Momjian br...@momjian.us writes: As you can see, ALTER INDEX renamed both the pg_constraint and pg_class names. Is it possible someone manually updated the system table to rename this primary key? That would cause this error message. The fix is to just to make sure they match. Does

Re: [HACKERS] Parser Cruft in gram.y

2012-12-20 Thread Andres Freund
On 2012-12-20 15:58:12 +, Greg Stark wrote: On Thu, Dec 20, 2012 at 3:18 AM, Tom Lane t...@sss.pgh.pa.us wrote: Greg Stark st...@mit.edu writes: But I'm not entirely convinced any of this is actually useful. Just becuase the transition table is large doesn't mean it's inefficient.

Re: [HACKERS] ThisTimeLineID in checkpointer and bgwriter processes

2012-12-20 Thread Fujii Masao
On Thu, Dec 20, 2012 at 8:41 PM, Heikki Linnakangas hlinnakan...@vmware.com wrote: On 20.12.2012 12:08, Amit Kapila wrote: On Wednesday, December 19, 2012 9:30 PM Heikki Linnakangas wrote: In both checkpointer.c and bgwriter.c, we do this before entering the main loop: /*

Re: [HACKERS] ThisTimeLineID in checkpointer and bgwriter processes

2012-12-20 Thread Tom Lane
Simon Riggs si...@2ndquadrant.com writes: PreallocXlogFiles() should have been put to the sword long ago. It's a performance tweak aimed at people without a performance problem in this area. This claim seems remarkably lacking in any supporting evidence. I'll freely grant that

Re: [HACKERS] operator dependency of commutator and negator, redux

2012-12-20 Thread Tom Lane
Robert Haas robertmh...@gmail.com writes: On Thu, Dec 20, 2012 at 10:52 AM, Tom Lane t...@sss.pgh.pa.us wrote: I was thinking a NOTICE at most. If it's a WARNING then restoring perfectly valid pg_dump files will result in lots of scary-looking chatter. You could make an argument for printing

[HACKERS] ALTER .. OWNER TO error mislabels schema as other object type

2012-12-20 Thread Robert Haas
This looks busted: rhaas=# create role clerks; CREATE ROLE rhaas=# create role bob in role clerks; CREATE ROLE rhaas=# create schema foo; CREATE SCHEMA rhaas=# grant usage on schema foo to bob, clerks; GRANT rhaas=# create aggregate foo.sum(basetype=text,sfunc=textcat,stype=text,initcond='');

Re: [HACKERS] Enabling Checksums

2012-12-20 Thread Martijn van Oosterhout
On Tue, Dec 18, 2012 at 04:06:02AM -0500, Greg Smith wrote: On 12/18/12 3:17 AM, Simon Riggs wrote: Clearly part of the response could involve pg_dump on the damaged structure, at some point. This is the main thing I wanted to try out more, once I have a decent corruption generation tool.

Re: [HACKERS] ThisTimeLineID in checkpointer and bgwriter processes

2012-12-20 Thread Simon Riggs
On 20 December 2012 13:19, Amit Kapila amit.kap...@huawei.com wrote: True, it might not have any functionality effect in RemoveOldXlogFiles(). However it can be used in PreallocXlogFiles()-XLogFileInit() as well. Which is never called in recovery because we never write WAL. -- Simon Riggs

Re: [HACKERS] ALTER .. OWNER TO error mislabels schema as other object type

2012-12-20 Thread Tom Lane
Robert Haas robertmh...@gmail.com writes: This looks busted: Between this and your previous example, it's becoming clear that the recent refactorings of the ALTER code were not ready for prime time. Perhaps we should just revert those instead of playing bug whack-a-mole.

Re: [HACKERS] Switching timeline over streaming replication

2012-12-20 Thread Fujii Masao
On Sat, Dec 15, 2012 at 9:36 AM, Fujii Masao masao.fu...@gmail.com wrote: On Sat, Dec 8, 2012 at 12:51 AM, Heikki Linnakangas hlinnakan...@vmware.com wrote: On 06.12.2012 15:39, Amit Kapila wrote: On Thursday, December 06, 2012 12:53 AM Heikki Linnakangas wrote: On 05.12.2012 14:32, Amit

Re: [HACKERS] Set visibility map bit after HOT prune

2012-12-20 Thread Pavan Deolasee
On Thu, Dec 20, 2012 at 9:23 PM, Robert Haas robertmh...@gmail.com wrote: On Wed, Dec 19, 2012 at 11:12 PM, Pavan Deolasee I'm very reluctant to suggest that we can solve this my setting aside another page-level bit to track visibility of tuples for heapscans. Or even have a bit in the tuple

Re: [HACKERS] ThisTimeLineID in checkpointer and bgwriter processes

2012-12-20 Thread Simon Riggs
On 20 December 2012 16:21, Tom Lane t...@sss.pgh.pa.us wrote: Simon Riggs si...@2ndquadrant.com writes: PreallocXlogFiles() should have been put to the sword long ago. It's a performance tweak aimed at people without a performance problem in this area. This claim seems remarkably lacking in

Re: [HACKERS] ThisTimeLineID in checkpointer and bgwriter processes

2012-12-20 Thread Fujii Masao
On Fri, Dec 21, 2012 at 1:46 AM, Simon Riggs si...@2ndquadrant.com wrote: On 20 December 2012 13:19, Amit Kapila amit.kap...@huawei.com wrote: True, it might not have any functionality effect in RemoveOldXlogFiles(). However it can be used in PreallocXlogFiles()-XLogFileInit() as well. Which

Re: [HACKERS] ThisTimeLineID in checkpointer and bgwriter processes

2012-12-20 Thread Andres Freund
On 2012-12-20 16:46:21 +, Simon Riggs wrote: On 20 December 2012 13:19, Amit Kapila amit.kap...@huawei.com wrote: True, it might not have any functionality effect in RemoveOldXlogFiles(). However it can be used in PreallocXlogFiles()-XLogFileInit() as well. Which is never called in

Re: [HACKERS] FDW: ForeignPlan and parameterized paths

2012-12-20 Thread Tom Lane
Ronan Dunklau rdunk...@gmail.com writes: I intentionally did the nestloop_params substitution after calling GetForeignPlan not before. It's not apparent to me why it would be useful to do it before, because the FDW is going to have no idea what those params represent. (Note that they

Re: [HACKERS] ALTER .. OWNER TO error mislabels schema as other object type

2012-12-20 Thread Robert Haas
On Thu, Dec 20, 2012 at 11:46 AM, Tom Lane t...@sss.pgh.pa.us wrote: Robert Haas robertmh...@gmail.com writes: This looks busted: Between this and your previous example, it's becoming clear that the recent refactorings of the ALTER code were not ready for prime time. Perhaps we should just

Re: [HACKERS] Set visibility map bit after HOT prune

2012-12-20 Thread Tom Lane
Pavan Deolasee pavan.deola...@gmail.com writes: Ok. Will try to read archives to see what was actually suggested and why it was put on back burner. BTW at the risk of being shot down again, I wonder if can we push down the freeze operation to HOT prune also. Seems unlikely to be a win. We

Re: [HACKERS] Set visibility map bit after HOT prune

2012-12-20 Thread Robert Haas
On Thu, Dec 20, 2012 at 11:49 AM, Pavan Deolasee pavan.deola...@gmail.com wrote: I wonder if we should add a flag to heap_page_prune and try to do some additional work if its being called from lazy vacuum such as setting the VM bit and the tuple freeze. IIRC I had put something like that in

Re: [HACKERS] ThisTimeLineID in checkpointer and bgwriter processes

2012-12-20 Thread Heikki Linnakangas
On 20.12.2012 18:19, Fujii Masao wrote: InstallXLogFileSegment() also uses ThisTimeLineID. But your recent commit doesn't take care of it and prevents the standby from recycling the WAL files properly. Specifically, the standby recycles the WAL file to wrong name. A-ha, good catch. So that's

Re: [HACKERS] Set visibility map bit after HOT prune

2012-12-20 Thread Pavan Deolasee
On Thu, Dec 20, 2012 at 10:55 PM, Tom Lane t...@sss.pgh.pa.us wrote: Pavan Deolasee pavan.deola...@gmail.com writes: Ok. Will try to read archives to see what was actually suggested and why it was put on back burner. BTW at the risk of being shot down again, I wonder if can we push down the

Re: [HACKERS] Set visibility map bit after HOT prune

2012-12-20 Thread Pavan Deolasee
On Thu, Dec 20, 2012 at 10:59 PM, Robert Haas robertmh...@gmail.com wrote: On Thu, Dec 20, 2012 at 11:49 AM, Pavan Deolasee pavan.deola...@gmail.com wrote: I wonder if we should add a flag to heap_page_prune and try to do some additional work if its being called from lazy vacuum such as

Re: [HACKERS] pg_top

2012-12-20 Thread John R Pierce
On 12/20/2012 4:17 AM, Brett Maton wrote: It appears that procpid has been renamed to pid at some point, also the column current_query appears to have been shortened to query. My patch updates a couple of queries to use the new shorter column names. IMHO, any such fix should check the

Re: [HACKERS] [GENERAL] trouble with pg_upgrade 9.0 - 9.1

2012-12-20 Thread Bruce Momjian
On Thu, Dec 20, 2012 at 11:08:58AM -0500, Tom Lane wrote: Bruce Momjian br...@momjian.us writes: As you can see, ALTER INDEX renamed both the pg_constraint and pg_class names. Is it possible someone manually updated the system table to rename this primary key? That would cause this error

Re: [HACKERS] Set visibility map bit after HOT prune

2012-12-20 Thread Tom Lane
Pavan Deolasee pavan.deola...@gmail.com writes: On Thu, Dec 20, 2012 at 10:55 PM, Tom Lane t...@sss.pgh.pa.us wrote: Seems unlikely to be a win. We only care about freezing tuples in the context of being able to advance a relation-wide relfrozenxid horizon. Freezing pages retail accomplishes

Re: [HACKERS] Set visibility map bit after HOT prune

2012-12-20 Thread Robert Haas
On Thu, Dec 20, 2012 at 1:30 PM, Tom Lane t...@sss.pgh.pa.us wrote: Pavan Deolasee pavan.deola...@gmail.com writes: On Thu, Dec 20, 2012 at 10:55 PM, Tom Lane t...@sss.pgh.pa.us wrote: Seems unlikely to be a win. We only care about freezing tuples in the context of being able to advance a

Re: [HACKERS] [GENERAL] trouble with pg_upgrade 9.0 - 9.1

2012-12-20 Thread Bruce Momjian
Applied. We can mark this report closed. Groshev, let us know if you have any further problems. --- On Thu, Dec 20, 2012 at 07:19:48AM -0500, Bruce Momjian wrote: On Wed, Dec 19, 2012 at 10:19:30PM -0500, Bruce Momjian

Re: [HACKERS] Feature Request: pg_replication_master()

2012-12-20 Thread Bruce Momjian
On Wed, Dec 19, 2012 at 07:32:33PM -0500, Robert Haas wrote: On Wed, Dec 19, 2012 at 5:34 PM, Simon Riggs si...@2ndquadrant.com wrote: As ever, we spent much energy on debating backwards compatibility rather than just solving the problem it posed, which is fairly easy to solve. I'm still

Re: [HACKERS] Feature Request: pg_replication_master()

2012-12-20 Thread Bruce Momjian
On Wed, Dec 19, 2012 at 10:34:14PM +, Simon Riggs wrote: On 19 December 2012 22:19, Joshua Berkus j...@agliodbs.com wrote: It stalled because the patch author decided not to implement the request to detect recovery.conf in data directory, which allows backwards compatibility.

Re: [HACKERS] Review of Row Level Security

2012-12-20 Thread Stephen Frost
Kevin, all, * Kevin Grittner (kgri...@mail.com) wrote: The more secure behavior is to allow entry of data which will not be visible by the person doing the entry. wrt this- I'm inclined to agree with Kevin. It's certainly common in certain environments that you can write to a higher level

Re: [HACKERS] Review of Row Level Security

2012-12-20 Thread Robert Haas
On Thu, Dec 20, 2012 at 4:35 AM, Simon Riggs si...@2ndquadrant.com wrote: Not sure I understand you. You suggested it was a valid use case for a user to have only INSERT privilege and wish to bypass security checks. I agreed and suggested it could be special-cased. That's not really what I

Re: [HACKERS] Feature Request: pg_replication_master()

2012-12-20 Thread Bruce Momjian
On Thu, Dec 20, 2012 at 02:29:49PM -0500, Bruce Momjian wrote: Let me also add that I am tired of having recovery.conf improvement stalled by backward compatibility concerns. At this point, let's just trash recovery.conf backward compatibility and move on. And I don't want to hear

Re: [HACKERS] Review of Row Level Security

2012-12-20 Thread Kohei KaiGai
2012/12/20 Stephen Frost sfr...@snowman.net: Kevin, all, * Kevin Grittner (kgri...@mail.com) wrote: The more secure behavior is to allow entry of data which will not be visible by the person doing the entry. wrt this- I'm inclined to agree with Kevin. It's certainly common in certain

Re: [HACKERS] Review of Row Level Security

2012-12-20 Thread Stephen Frost
* Robert Haas (robertmh...@gmail.com) wrote: * Applies to all commands should not be implemented via triggers. Complex, slow, unacceptable thing to force upon users. Doing that begs the question of why we would have the feature at all, since we already have triggers and barrier views. I

Re: [HACKERS] Review of Row Level Security

2012-12-20 Thread Kohei KaiGai
2012/12/20 Robert Haas robertmh...@gmail.com: On Thu, Dec 20, 2012 at 4:35 AM, Simon Riggs si...@2ndquadrant.com wrote: Not sure I understand you. You suggested it was a valid use case for a user to have only INSERT privilege and wish to bypass security checks. I agreed and suggested it could

Re: [HACKERS] Feature Request: pg_replication_master()

2012-12-20 Thread Petr Jelinek
Let me also add that I am tired of having recovery.conf improvement stalled by backward compatibility concerns. At this point, let's just trash recovery.conf backward compatibility and move on. And I don't want to hear complaints about tool breakage either. These are external tools, not

Re: [HACKERS] operator dependency of commutator and negator, redux

2012-12-20 Thread Dimitri Fontaine
Robert Haas robertmh...@gmail.com writes: a separate ALTER OPERATOR COMMUTATOR statement (or something of the sort) that pg_dump can emit as a separate item. Even a NOTICE in I like that capability, but it's not helping us in the backward compatibility section where we will still read

Re: [HACKERS] strange OOM errors with EXECUTE in PL/pgSQL

2012-12-20 Thread Dimitri Fontaine
Tom Lane t...@sss.pgh.pa.us writes: The reason this fails is that you've got a half-megabyte source string, and each of the 11000 plans that are due to be created from it saves its own copy of the source string. Hence, 5500 megabytes needed just for source strings. We could possibly fix

Re: [HACKERS] Review of Row Level Security

2012-12-20 Thread Kevin Grittner
Kohei KaiGai wrote: If system ensures writer's permission is always equivalent or more restrictive than reader's permission, it also eliminates the problem around asymmetric row-security policy between commands. I'm not sure we're understanding each other; so far people who favor asymmetric

Re: [HACKERS] Feature Request: pg_replication_master()

2012-12-20 Thread Joshua Berkus
As ever, we spent much energy on debating backwards compatibility rather than just solving the problem it posed, which is fairly easy to solve. Well, IIRC, the debate was primarily of *your* making. Almost everyone else on the thread was fine with the original patch, and it was nearly done

Re: [HACKERS] Feature Request: pg_replication_master()

2012-12-20 Thread Andres Freund
On 2012-12-18 19:43:09 -0800, Josh Berkus wrote: We should probably have a function, like pg_replication_master(), which gives the host address of the current master. This would help DBAs for large replication clusters a lot. Obviously, this would only work in streaming. Do you want the

Re: [HACKERS] Switching timeline over streaming replication

2012-12-20 Thread Joshua Berkus
I just committed a patch that should make the requested WAL segment 00020003 has already been removed errors go away. The trick was for walsenders to not switch to the new timeline until at least one record has been replayed on it. That closes the window where the

Re: [HACKERS] pg_top

2012-12-20 Thread Tom Lane
John R Pierce pie...@hogranch.com writes: /me tossed another mumbled curse at whomever changed that field name. The reason for the field name change was that the semantics of the field changed. You typically ought to look at what the application is actually doing with the field, not just do

Re: [HACKERS] pg_top

2012-12-20 Thread Merlin Moncure
On Thu, Dec 20, 2012 at 4:17 PM, Tom Lane t...@sss.pgh.pa.us wrote: John R Pierce pie...@hogranch.com writes: /me tossed another mumbled curse at whomever changed that field name. The reason for the field name change was that the semantics of the field changed. You typically ought to look at

Re: [HACKERS] Cascading replication: should we detect/prevent cycles?

2012-12-20 Thread Joshua Berkus
Robert, What would such a test look like? It's not obvious to me that there's any rapid way for a user to detect this situation, without checking each server individually. Change something on the master and observe that none of the supposed standbys notice? That doesn't sound like an

Re: [HACKERS] Feature Request: pg_replication_master()

2012-12-20 Thread Joshua Berkus
Andreas, Do you want the node one step up or the top-level in the chain? Because I don't think we can do the latter without complicating the replication protocol noticably. Well, clearly a whole chain would be nice for the user. But even just one step up would be very useful. --Josh

Re: [HACKERS] Parser Cruft in gram.y

2012-12-20 Thread McDevitt, Charles
Another way of attack along these lines might be to use the %glr-parser and then try to cut back on all those redundant rules that were put in to avoid conflicts. The number of key words categories and such could perhaps also be reduced that way. Of course, this is mostly speculation.

Re: [HACKERS] Parser Cruft in gram.y

2012-12-20 Thread Andres Freund
On 2012-12-20 23:12:55 +, McDevitt, Charles wrote: Another way of attack along these lines might be to use the %glr-parser and then try to cut back on all those redundant rules that were put in to avoid conflicts. The number of key words categories and such could perhaps also be

Re: [HACKERS] Parser Cruft in gram.y

2012-12-20 Thread McDevitt, Charles
The GLR output from Bison is licensed under the GPL (unlike the LALR output). So using Bison's GLR mode isn't an option. Thats not the case anymore: http://www.gnu.org/software/bison/manual/html_node/Conditions.html Sorry! My mistake... I didn't realize they changed the rules. I'll

Re: [HACKERS] ThisTimeLineID in checkpointer and bgwriter processes

2012-12-20 Thread Fujii Masao
On Fri, Dec 21, 2012 at 2:45 AM, Heikki Linnakangas hlinnakan...@vmware.com wrote: On 20.12.2012 18:19, Fujii Masao wrote: InstallXLogFileSegment() also uses ThisTimeLineID. But your recent commit doesn't take care of it and prevents the standby from recycling the WAL files properly.

Re: [HACKERS] Parser Cruft in gram.y

2012-12-20 Thread Tom Lane
Andres Freund and...@2ndquadrant.com writes: On 2012-12-20 23:12:55 +, McDevitt, Charles wrote: Another way of attack along these lines might be to use the %glr-parser and then try to cut back on all those redundant rules that were put in to avoid conflicts. The number of key words

Re: [HACKERS] Switching timeline over streaming replication

2012-12-20 Thread Thom Brown
On 20 December 2012 12:45, Heikki Linnakangas hlinnakan...@vmware.comwrote: On 17.12.2012 15:05, Thom Brown wrote: I just set up 120 chained standbys, and for some reason I'm seeing these errors: LOG: replication terminated by primary server DETAIL: End of WAL reached on timeline 1 LOG:

Re: [HACKERS] Review of Row Level Security

2012-12-20 Thread Simon Riggs
On 20 December 2012 21:50, Kevin Grittner kgri...@mail.com wrote: How about using existing GRANT syntax but allowing a WHERE clause? It's a nice feature, but a completely different thing to what is being discussed here. Row security adds the ability to enforce a single coherent policy at

Re: [HACKERS] Set visibility map bit after HOT prune

2012-12-20 Thread Jeff Janes
On Wednesday, December 19, 2012, Robert Haas wrote: On Wed, Dec 19, 2012 at 12:26 PM, Pavan Deolasee pavan.deola...@gmail.com javascript:; wrote: I would like to run some pgbench tests where we get the system in a steady state such as all/most updates are HOT updates (not entirely

[HACKERS] Set visibility map bit after HOT prune

2012-12-20 Thread Jeff Janes
On Thursday, December 20, 2012, Robert Haas wrote: IMHO, it's probably fairly hopeless to make a pure pgbench workload show a benefit from index-only scans. A large table under a very heavy, completely random write workload is just about the worst possible case for index-only scans.

Re: [HACKERS] [PERFORM] pgbench to the MAXINT

2012-12-20 Thread Gurjeet Singh
On Wed, Feb 16, 2011 at 8:15 AM, Greg Smith g...@2ndquadrant.com wrote: Tom Lane wrote: I think that might be a good idea --- it'd reduce the cross-platform variability of the results quite a bit, I suspect. random() is not to be trusted everywhere, but I think erand48 is pretty much the

Re: [HACKERS] ThisTimeLineID in checkpointer and bgwriter processes

2012-12-20 Thread Amit Kapila
On Thursday, December 20, 2012 11:15 PM Heikki Linnakangas wrote: On 20.12.2012 18:19, Fujii Masao wrote: InstallXLogFileSegment() also uses ThisTimeLineID. But your recent commit doesn't take care of it and prevents the standby from recycling the WAL files properly. Specifically, the

Re: [HACKERS] Review of Row Level Security

2012-12-20 Thread Kohei KaiGai
2012/12/20 Kevin Grittner kgri...@mail.com: Kohei KaiGai wrote: If system ensures writer's permission is always equivalent or more restrictive than reader's permission, it also eliminates the problem around asymmetric row-security policy between commands. I'm not sure we're understanding

  1   2   >