Re: [HACKERS] autonomous transactions

2016-10-11 Thread Petr Jelinek
On 10/10/16 16:44, Merlin Moncure wrote: > On Thu, Oct 6, 2016 at 3:53 PM, Simon Riggs wrote: >> On 6 October 2016 at 21:27, Robert Haas wrote: >>> I think we should implement background transactions and call them >>> background transactions. That

Re: [HACKERS] autonomous transactions

2016-10-11 Thread Pavel Stehule
2016-10-11 17:06 GMT+02:00 Petr Jelinek : > On 10/10/16 16:44, Merlin Moncure wrote: > > On Thu, Oct 6, 2016 at 3:53 PM, Simon Riggs > wrote: > >> On 6 October 2016 at 21:27, Robert Haas wrote: > >>> I think we should implement

Re: [HACKERS] int2vector and btree indexes

2016-10-11 Thread Tom Lane
Amit Langote writes: > I was wrong that the index *never* gets used. It does in fact get used if > the operator is an ordering search operator (<, <=, >, >=), in which case > the query would use an array_ops operator (which is a btree operator class > for type

[HACKERS] Proposal: scan key push down to heap [WIP]

2016-10-11 Thread Dilip Kumar
Hi Hackers, I would like to propose a patch for pushing down the scan key to heap. Currently only in case of system table scan keys are pushed down. I have implemented the POC patch to do the same for normal table scan. This patch will extract the expression from qual and prepare the scan keys.

Re: [HACKERS] autonomous transactions

2016-10-11 Thread Merlin Moncure
On Tue, Oct 11, 2016 at 10:06 AM, Petr Jelinek wrote: > On 10/10/16 16:44, Merlin Moncure wrote: >> On Thu, Oct 6, 2016 at 3:53 PM, Simon Riggs wrote: >>> On 6 October 2016 at 21:27, Robert Haas wrote: I think we should

Re: [HACKERS] Is it time to kill support for very old servers?

2016-10-11 Thread Heikki Linnakangas
On 10/11/2016 08:23 PM, Tom Lane wrote: Not sure where to go from here, but the idea of dropping V2 support is seeming attractive again. Or we could just continue the policy of benign neglect. Let's drop it. - Heikki -- Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org) To

Re: [HACKERS] [PATCH] pgpassfile connection option

2016-10-11 Thread Oskari Saarenmaa
05.10.2016, 20:06, Robert Haas kirjoitti: You could do something like that, I guess, but I think it might be a good idea to wait and see if anyone else has opinions on (1) the desirability of the basic feature, (2) the severity of the security hazard it creates, and (3) your proposed remediation

Re: [HACKERS] Is it time to kill support for very old servers?

2016-10-11 Thread Tom Lane
Robert Haas writes: > On Fri, Oct 7, 2016 at 11:34 AM, Tom Lane wrote: >> The problem with letting it just sit there is that we're not, in fact, >> testing it. If we take the above argument seriously then we should >> provide some way to configure

Re: [HACKERS] Macro customizable hashtable / bitmapscan & aggregation perf

2016-10-11 Thread Andres Freund
On 2016-10-11 04:29:31 +0200, Tomas Vondra wrote: > On 10/11/2016 04:07 AM, Andres Freund wrote: > > On 2016-10-10 17:46:22 -0700, Andres Freund wrote: > > > > TPC-DS (tpcds.ods) > > > > -- > > > > > > > > In this case, I'd say the results are less convincing. There are quite > >

Re: [HACKERS] Merge Join with an Index Optimization

2016-10-11 Thread Claudio Freire
On Sun, Sep 11, 2016 at 1:20 PM, Tom Lane wrote: > Michael Malis writes: >> As I understand it, a merge join will currently read all tuples from both >> subqueries (besides early termination). I believe it should be possible to >> take advantages of

Re: [HACKERS] Is it time to kill support for very old servers?

2016-10-11 Thread Tom Lane
I wrote: > pg_dump alleges support for dumping from servers back to 7.0. Would v10 > be a good time to remove some of that code? It's getting harder and > harder to even compile those ancient branches, let alone get people to > test against them (cf. 4806f26f9). My initial thought is to cut

[HACKERS] buildfarm client release 4.18

2016-10-11 Thread Andrew Dunstan
I have just released buildfarm client version 4.18. In addition to some minor fixes, there are two significant changes: * The client now makes a determined effort to clean up any left over build artefacts from previous runs at the start of a run. It also tries to clean away old socket

Re: [HACKERS] Remove "Source Code" column from \df+ ?

2016-10-11 Thread Tom Lane
Stephen Frost writes: > As was mentioned, this thread doesn't really need a patch but rather > some comment from those who have voiced a -1 on removing the PL source > code column. > In another, perhaps vain, attempt to get to a consensus, here's what it > looks like the

Re: [HACKERS] DISTINCT with btree skip scan

2016-10-11 Thread Thomas Munro
On Sat, Nov 1, 2014 at 10:35 AM, Thomas Munro wrote: > I am definitely interested in collaborating on a series of patches to > implement various kinds of skip-based plans as seen in other RDBMSs if > others think it could be useful. I see skip-based DISTINCT as a good > place to

Re: [HACKERS] autonomous transactions

2016-10-11 Thread Pavel Stehule
2016-10-11 21:54 GMT+02:00 Merlin Moncure : > On Tue, Oct 11, 2016 at 10:06 AM, Petr Jelinek > wrote: > > On 10/10/16 16:44, Merlin Moncure wrote: > >> On Thu, Oct 6, 2016 at 3:53 PM, Simon Riggs > wrote: > >>> On 6 October 2016

Re: [HACKERS] int2vector and btree indexes

2016-10-11 Thread Amit Langote
On 2016/10/11 21:40, Tom Lane wrote: > Amit Langote writes: >> I was wrong that the index *never* gets used. It does in fact get used if >> the operator is an ordering search operator (<, <=, >, >=), in which case >> the query would use an array_ops operator (which

Re: [HACKERS] New SQL counter statistics view (pg_stat_sql)

2016-10-11 Thread Haribabu Kommi
On Thu, Sep 29, 2016 at 3:45 PM, Haribabu Kommi wrote: > > > On Thu, Sep 22, 2016 at 3:05 AM, Alvaro Herrera > wrote: > >> Peter Eisentraut wrote: >> >> > How about having the tag not be a column name but a row entry. So you'd >> > do

Re: [HACKERS] Remove "Source Code" column from \df+ ?

2016-10-11 Thread Pavel Stehule
2016-10-12 1:51 GMT+02:00 Tom Lane : > Stephen Frost writes: > > As was mentioned, this thread doesn't really need a patch but rather > > some comment from those who have voiced a -1 on removing the PL source > > code column. > > > In another, perhaps

Re: [HACKERS] New SQL counter statistics view (pg_stat_sql)

2016-10-11 Thread vinayak
On 2016/10/12 12:21, Haribabu Kommi wrote: On Thu, Sep 29, 2016 at 3:45 PM, Haribabu Kommi > wrote: On Thu, Sep 22, 2016 at 3:05 AM, Alvaro Herrera > wrote:

Re: [HACKERS] "Some tests to cover hash_index"

2016-10-11 Thread Mithun Cy
On Wed, Sep 28, 2016 at 9:56 PM, Robert Haas wrote: > something committable will come from it, but with 2 days left it's not > going to happen this CF. Adding a new patch. This one uses generate series instead of INSERT INTO SELECT and fixed comments from Alvaro. --

Re: [HACKERS] How to inspect tuples during execution of a plan?

2016-10-11 Thread Ashutosh Bapat
On Mon, Oct 10, 2016 at 1:39 PM, Ernst-Georg Schmid wrote: > Hello all, > > I'd like to inspect the content of tuples as they are sent during the > execution of a query in order to react to their values. The correct answer will depend upon the purpose of this

[HACKERS] Remove "Source Code" column from \df+ ?

2016-10-11 Thread Stephen Frost
All, Starting a new thread with an accurate name to see if we can't get somewhere with this topic. * Pavel Stehule (pavel.steh...@gmail.com) wrote: > 2016-10-08 23:46 GMT+02:00 Jim Nasby : > > On 10/3/16 3:18 PM, Pavel Stehule wrote: > >> I am feeling consensus on

Re: [HACKERS] Macro customizable hashtable / bitmapscan & aggregation perf

2016-10-11 Thread Tomas Vondra
On 10/11/2016 05:56 PM, Andres Freund wrote: On 2016-10-11 04:29:31 +0200, Tomas Vondra wrote: On 10/11/2016 04:07 AM, Andres Freund wrote: On 2016-10-10 17:46:22 -0700, Andres Freund wrote: TPC-DS (tpcds.ods) -- In this case, I'd say the results are less convincing. There

Re: [HACKERS] Forbid use of LF and CR characters in database and role names

2016-10-11 Thread Michael Paquier
On Tue, Oct 11, 2016 at 11:00 AM, Noah Misch wrote: > I count one person disfavoring the patch concept of rejecting these characters > early, and I count two people, plus yourself as author, favoring it. > Therefore, the patch can move forward with the proposed design. The

[HACKERS] int2vector and btree indexes

2016-10-11 Thread Amit Langote
If I create btree index on a int2vector column, it does not get used for queries because the query search always fails to match the index operator (family). During index creation, GetDefaultOpClass() returns array_ops for a int2vector index column, because type int2vector is binary-coercible with

Re: [HACKERS] proposal: psql \setfileref

2016-10-11 Thread Pavel Stehule
2016-10-11 9:32 GMT+02:00 Gilles Darold : > Le 11/10/2016 à 07:53, Pavel Stehule a écrit : > > > > 2016-10-10 19:50 GMT+02:00 Pavel Stehule : > >> >> >> 2016-10-10 15:17 GMT+02:00 Gilles Darold : >> >>> Le 10/10/2016 à

Re: [HACKERS] proposal: psql \setfileref

2016-10-11 Thread Gilles Darold
Le 11/10/2016 à 07:53, Pavel Stehule a écrit : > > > 2016-10-10 19:50 GMT+02:00 Pavel Stehule >: > > > > 2016-10-10 15:17 GMT+02:00 Gilles Darold >: > > Le

Re: [HACKERS] kqueue

2016-10-11 Thread Torsten Zuehlsdorff
On 28.09.2016 23:39, Thomas Munro wrote: On Thu, Sep 29, 2016 at 9:09 AM, Keith Fiske wrote: On Thu, Sep 15, 2016 at 11:11 PM, Thomas Munro wrote: Ok, here's a version tweaked to use EVFILT_PROC for postmaster death detection instead of the

Re: [HACKERS] int2vector and btree indexes

2016-10-11 Thread Amit Langote
On 2016/10/11 15:58, Amit Langote wrote: > If I create btree index on a int2vector column, it does not get used for > queries because the query search always fails to match the index operator > (family). > > During index creation, GetDefaultOpClass() returns array_ops for a > int2vector index

Re: [HACKERS] Quorum commit for multiple synchronous replication.

2016-10-11 Thread Masahiko Sawada
On Wed, Sep 28, 2016 at 5:14 PM, Michael Paquier wrote: > On Sat, Sep 24, 2016 at 5:37 PM, Masahiko Sawada > wrote: >> I still vote for changing behaviour of existing syntax 'k (n1, n2)' to >> quorum commit. >> That is, >> 1. 'First k (n1, n2,

Re: [HACKERS] proposal: schema PL session variables

2016-10-11 Thread Pavel Stehule
Hi 2016-02-23 20:52 GMT+01:00 Pavel Stehule : > > > 2016-02-12 22:41 GMT+01:00 Jim Nasby : > >> On 2/12/16 2:58 PM, Pavel Stehule wrote: >> >>> >>> So I think adding something like this needs to at least address >>> *how* SQL level

Re: [HACKERS] memory leak in e94568ecc10 (pre-reading in external sort)

2016-10-11 Thread Heikki Linnakangas
On 10/11/2016 12:56 AM, Peter Geoghegan wrote: Also, something about trace_sort here: +#ifdef TRACE_SORT + if (trace_sort) + elog(LOG, "using " INT64_FORMAT " KB of memory for read buffers among %d input tapes", +(state->availMem) / 1024, numInputTapes); +#endif + +

Re: [HACKERS] Quorum commit for multiple synchronous replication.

2016-10-11 Thread Michael Paquier
On Tue, Oct 11, 2016 at 4:18 PM, Masahiko Sawada wrote: >> You may want to add an assert in >> SyncRepGetSyncStandbysPriority and SyncRepGetSyncStandbysQuorum to be >> sure that they are getting called for the correct method. >> + /* Sort each array in descending order to

Re: [HACKERS] Quorum commit for multiple synchronous replication.

2016-10-11 Thread Michael Paquier
On Tue, Oct 11, 2016 at 6:08 PM, Michael Paquier wrote: > On Tue, Oct 11, 2016 at 4:18 PM, Masahiko Sawada > wrote: >>> You may want to add an assert in >>> SyncRepGetSyncStandbysPriority and SyncRepGetSyncStandbysQuorum to be >>> sure that they