On Fri, Mar 17, 2017 at 9:03 AM, Amit Kapila wrote:
> On Thu, Mar 16, 2017 at 1:15 PM, Ashutosh Sharma
> wrote:
>> Hi,
>>
>> Attached is the patch that allows WAL consistency tool to mask
>> 'LH_PAGE_HAS_DEAD_TUPLES' flag in hash index. The flag got added as a
>> part of 'Microvacuum support for
>
> Nikhil, do you mind if I try something like that? As we already know
> what is the first XID when beginning redo via
> ShmemVariableCache->nextXid it is possible to discard 2PC files that
> should not be here.
Yeah, that is ok.
> What makes me worry is the control of the maximum
> number of
>
> >
> > I don't think this will work. We cannot replace pg_twophase with shmem
> > entries + WAL pointers. This is because we cannot expect to have WAL
> entries
> > around for long running prepared queries which survive across
> checkpoints.
>
> But at the beginning of recovery, we can mark such
Hello,
At Mon, 13 Mar 2017 11:06:00 +1100, Venkata B Nagothi wrote
in
> On Tue, Jan 17, 2017 at 9:36 PM, Kyotaro HORIGUCHI <
> horiguchi.kyot...@lab.ntt.co.jp> wrote:
> > I managed to reproduce this. A little tweak as the first patch
> > lets the standby to suicide as soon as walreceiver sees a
On Fri, Mar 17, 2017 at 4:42 PM, Nikhil Sontakke
wrote:
>> >
>> > I don't think this will work. We cannot replace pg_twophase with shmem
>> > entries + WAL pointers. This is because we cannot expect to have WAL
>> > entries
>> > around for long running prepared queries which survive across
>> > ch
At Mon, 13 Mar 2017 10:42:05 -0400, David Steele wrote in
<1e8297fd-f7f2-feab-848d-5121e45c8...@pgmasters.net>
> It has been a while since this thread has received any comments or a new
> patch. The general consensus seems to be that this feature is too large
> a rewrite of tab completion consid
>
> >
> > Ok, we can do that and then yes, RecoverPreparedTransaction() can just
> have
> > one loop going through the shmem entries. BUT, we cannot ignore
> > "inredo"+"ondisk" entries. For such entries, we will have to read and
> > recover from the corresponding 2PC files.
>
> Yes. About other th
From: pgsql-hackers-ow...@postgresql.org
> [mailto:pgsql-hackers-ow...@postgresql.org] On Behalf Of Ashutosh Bapat
> The scope of this work has expanded, since last time I reviewed and marked
> it as RFC. Right now I am busy with partition-wise joins and do not have
> sufficient time to take a look
On Fri, Mar 17, 2017 at 5:00 PM, Nikhil Sontakke
wrote:
>> >
>> > Ok, we can do that and then yes, RecoverPreparedTransaction() can just
>> > have
>> > one loop going through the shmem entries. BUT, we cannot ignore
>> > "inredo"+"ondisk" entries. For such entries, we will have to read and
>> > re
From: pgsql-hackers-ow...@postgresql.org
> [mailto:pgsql-hackers-ow...@postgresql.org] On Behalf Of Tsunakawa,
> Takayuki
> I made this ready for committer. The patch applied except for catversion.h,
> the patch content looks good, and the target test passed as follows:
Sorry, I reverted this to
At Thu, 16 Mar 2017 17:16:32 -0400, Corey Huinker
wrote in
> On Thu, Mar 16, 2017 at 4:17 PM, Tom Lane wrote:
>
> > Corey Huinker writes:
> > > I reworked the test such that all of the foreign tables inherit from the
> > > same parent table, and if you query that you do get async execution. B
On 2017/03/16 22:16, Robert Haas wrote:
> On Thu, Mar 16, 2017 at 6:03 AM, Amit Langote
> wrote:
>> I think we'll need to store *somewhere* the mapping of which inh=false
>> partitioned table RTE is the child of which inh=true (IOW, parent)
>> partitioned table RTE.
>
> I mean, for the children y
On Thu, Mar 16, 2017 at 7:05 PM, Robert Haas wrote:
> On Thu, Mar 16, 2017 at 6:09 AM, Dave Page wrote:
>> Hmm, good point. Google seems to be saying there isn't one. Patch
>> updated as you suggest (and I've added back in a function declaration
>> that got lost in the rebasing of the last versio
On Tue, Mar 14, 2017 at 1:35 AM, Michael Meskes wrote:
>> Previous 002 patch lacked to add describing PREPARE TRANSACTION.
>> Attached updated 002 patch.
>
> I just committed both patches and a backport of the bug fix itself.
> Thanks again for finding and fixing.
Regression tests for sql/twophase
On Thu, Mar 16, 2017 at 1:18 PM, Michael Paquier
wrote:
> On Wed, Mar 15, 2017 at 9:14 PM, Kuntal Ghosh
> wrote:
>> I've attached the updated patches.
>
> Thanks for the new versions. This begins to look really clear.
Thanks again for the review.
> Having some activity really depends on the back
On 03/17/2017 12:21 AM, MauMau wrote:
From: Heikki Linnakangas
So, I think we still need the check for Local System.
Thanks, fixed and confirmed that the error message is output in the
event log.
Committed, thanks!
- Heikki
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.o
On 03/14/2017 03:43 PM, Michael Paquier wrote:
+ /*
+* The password looked like a SCRAM verifier, but could not be
+* parsed.
+*/
+ elog(LOG, "invalid SCRAM verifier for user \"%s\"", username);
This would be sent back to the client, no? I think that you should
On 2017/03/17 10:38, Robert Haas wrote:
On Fri, Mar 10, 2017 at 2:46 AM, vinayak
wrote:
Thank you for reviewing the patch.
The attached patch incorporated Michael and Amit comments also.
I reviewed this tonight.
Thank you for reviewing the patch.
+/* Report compute index stats phas
Hi
On Thu, Mar 16, 2017 at 7:04 PM, Denish Patel wrote:
> Hi Dave,
>
> The patch failed applied...
>
> patch -p1 < /home/vagrant/pg_monitor.diff
> patching file contrib/pg_buffercache/Makefile
> patching file contrib/pg_buffercache/pg_buffercache--1.2--1.3.sql
> patching file contrib/pg_buffercac
On 2017-03-17 02:28, Corey Huinker wrote:
Attached is the latest work. Not everything is done yet. I post it
because
0001.if_endif.v23.diff
This patch does not compile for me (gcc 6.3.0):
command.c:38:25: fatal error: conditional.h: No such file or directory
#include "conditional.h"
Hi,
Am Freitag, den 17.03.2017, 10:50 +0900 schrieb Michael Paquier:
> The comment block format is incorrect. I would think as well that this
> comment should say it is important to have the main tablespace listed
> last it includes the WAL segments, and those need to contain all the
> latest WAL
I haven't yet handled all points, but meanwhile, some of the important
points are discussed below ...
On 6 March 2017 at 15:11, Amit Langote wrote:
>
>>> But that starts to sound less attractive when one realizes that
>>> that will occur for every row that wants to move.
>>
>> If we manage to cal
> 1.
>
> + Assert(nnumbers == 1);
>
> I think its a bad idea to Assert() this. The stat tuple can come from
> a plugin which could do anything. Seems like if we need to be certain
> of that then it should be an elog(ERROR), maybe mention that we
> expected a 1 element array, but got elements.
But
On 03/17/2017 07:19 AM, Kyotaro HORIGUCHI wrote:
At Mon, 13 Mar 2017 21:07:39 +0200, Heikki Linnakangas wrote in
Hmm. A somewhat different approach might be more suitable for testing
across versions, though. We could modify the perl scripts slightly to
print out SQL statements that exercise e
On Fri, Mar 17, 2017 at 12:17 PM, Kuntal Ghosh
wrote:
> On Tue, Mar 14, 2017 at 1:35 AM, Michael Meskes wrote:
>>> Previous 002 patch lacked to add describing PREPARE TRANSACTION.
>>> Attached updated 002 patch.
>>
>> I just committed both patches and a backport of the bug fix itself.
>> Thanks a
Tom Lane wrote:
> OT_WHOLE_LINE is not what you want because that results in verbatim
> copying, without variable expansion or anything
But if we want to implement "\if defined :foo" in the future
isn't it just what we need?
Also we could leave open the option to accept an SQL expression
On Fri, Mar 17, 2017 at 6:40 PM, Heikki Linnakangas wrote:
> Committed, thanks.
Thanks for the commit.
--
Michael
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers
On Fri, Mar 17, 2017 at 4:34 PM, Masahiko Sawada wrote:
> On Fri, Mar 17, 2017 at 12:17 PM, Kuntal Ghosh
> wrote:
>> On Tue, Mar 14, 2017 at 1:35 AM, Michael Meskes
>> wrote:
Previous 002 patch lacked to add describing PREPARE TRANSACTION.
Attached updated 002 patch.
>>>
>>> I just co
On Fri, Mar 17, 2017 at 7:18 PM, Michael Banck
wrote:
> Hi,
>
> Am Freitag, den 17.03.2017, 10:50 +0900 schrieb Michael Paquier:
>> The comment block format is incorrect. I would think as well that this
>> comment should say it is important to have the main tablespace listed
>> last it includes th
On Fri, Mar 17, 2017 at 2:08 AM, Beena Emerson wrote:
> The option was intended to only accept values in MB as the original config
> --with-wal-segsize option, unfortunately, the patch does not throw error as
> in the config option when the units are specified.
Yeah, you want to use strtol(), so
On Tue, 14 Mar 2017 10:08:14 -0400
David Steele wrote:
> Please post an explanation for the delay and a schedule for the new
> patch. If no patch or explanation is posted by 2017-03-17 AoE I will
> mark this submission "Returned with Feedback".
I am sorry for my late response. I had not a enoug
On Thu, Mar 16, 2017 at 11:38 PM, Michael Paquier
wrote:
> On Thu, Mar 16, 2017 at 10:52 PM, Heikki Linnakangas wrote:
>> On 03/14/2017 11:14 PM, Tom Lane wrote:
>>>
>>> In short, I don't think that argument refutes my position that "md5"
>>> in pg_hba.conf should be understood as allowing SCRAM
On Fri, Mar 17, 2017 at 12:30 PM, Ashutosh Sharma wrote:
> On Fri, Mar 17, 2017 at 9:03 AM, Amit Kapila wrote:
>> On Thu, Mar 16, 2017 at 1:15 PM, Ashutosh Sharma
>> wrote:
>>> Hi,
>>>
>>> Attached is the patch that allows WAL consistency tool to mask
>>> 'LH_PAGE_HAS_DEAD_TUPLES' flag in hash
On 03/17/2017 02:01 PM, Robert Haas wrote:
On Thu, Mar 16, 2017 at 11:38 PM, Michael Paquier
wrote:
At least this has the merit of making \password simpler from psql
without a kind of --method option: if the backend is 9.6 or older,
just generate a MD5-hash, and SCRAM-hash for newer versions.
P
On 3/17/17 2:08 AM, Fabien COELHO wrote:
>
> Hello David,
>
>>> Repost from bugs.
>>
>> This patch does not apply at cccbdde:
>
> Indeed. It should not. The fix is for the 9.6 branch. The issue has been
> fixed by some heavy but very welcome restructuring in master.
Whoops, sorry about that!
>
On Fri, Mar 17, 2017 at 12:27 PM, Ashutosh Sharma wrote:
> On Fri, Mar 17, 2017 at 8:20 AM, Amit Kapila wrote:
>
> As I said in my previous e-mail, I think you need
>> to record clearing of this flag in WAL record XLOG_HASH_DELETE as you
>> are not doing this unconditionally and then during repla
On Fri, Mar 17, 2017 at 8:32 AM, Heikki Linnakangas wrote:
> It would make sense to have \password obey password_encryption GUC. Then
> \password and ALTER USER would do the same thing, which would be less
> surprising. Although it's also a bit weird for a GUC to affect client-side
> behavior, so
"Daniel Verite" writes:
> Tom Lane wrote:
>> OT_WHOLE_LINE is not what you want because that results in verbatim
>> copying, without variable expansion or anything
> But if we want to implement "\if defined :foo" in the future
> isn't it just what we need?
I don't think that should mean wh
On Thu, Mar 16, 2017 at 6:30 PM, David Christensen wrote:
> The HA docs reference a “glossary” link which is no longer accessible, nor is
> it likely to be useful in general to link off-site IMHO. This simple patch
> removes this link.
Committed and back-patched.
Thanks.
--
Robert Haas
Ente
On 3/17/17 3:58 AM, Kyotaro HORIGUCHI wrote:
> At Mon, 13 Mar 2017 10:42:05 -0400, David Steele wrote
> in <1e8297fd-f7f2-feab-848d-5121e45c8...@pgmasters.net>
>> It has been a while since this thread has received any comments or a new
>> patch. The general consensus seems to be that this featur
On 3/13/17 11:15 AM, David Steele wrote:
> Hi Matheus,
>
> On 3/2/17 8:27 AM, David Steele wrote:
>> On 1/18/17 7:18 PM, Petr Jelinek wrote:
>>>
>>> The patch looks good, the only thing I am missing is tab completion
>>> support for psql.
>>
>> It looks like this patch is still waiting on an updat
Robert Haas writes:
> On Fri, Mar 17, 2017 at 8:32 AM, Heikki Linnakangas wrote:
>> It would make sense to have \password obey password_encryption GUC. Then
>> \password and ALTER USER would do the same thing, which would be less
>> surprising. Although it's also a bit weird for a GUC to affect c
On 3/15/17 1:56 AM, Tsunakawa, Takayuki wrote:
> From: pgsql-hackers-ow...@postgresql.org
>> [mailto:pgsql-hackers-ow...@postgresql.org] On Behalf Of David Steele
>> Sure, but having the private key may allow them to get new data from the
>> server as well as the data from the backup.
>
> You are
On 3/15/17 3:00 AM, Tsunakawa, Takayuki wrote:
> From: pgsql-hackers-ow...@postgresql.org
>> [mailto:pgsql-hackers-ow...@postgresql.org] On Behalf Of David Steele
>>> But it might be worth thinking about whether we want to encourage
>>> people to do manual chmod's at all; that's fairly easy to get
I've found various sources that give hints about what a student
proposal should look like, but nothing I could just give as a link,
so I pulled together what I could find, tempered by my own ideas and
opinions. I suggest that we send the below, or something like it to
each student who expresses in
On Fri, Mar 17, 2017 at 10:58 PM, Robert Haas wrote:
> Fine! I've committed the pg_clog renaming, but I'd really like to
> draw the line here. I'm not going to commit the pg_subtrans ->
> pg_subxact naming and am -1 on anyone else doing so. I think that
> having the names of things in the code
On Thu, Mar 16, 2017 at 10:21 PM, Michael Paquier
wrote:
> On Fri, Mar 17, 2017 at 11:17 AM, Robert Haas wrote:
>> I understand that the point of renaming pg_clog to pg_xact is that
>> pg_clog contains the dreaded letters l-o-g, which we hypothesize
>> causes DBAs to remove it. (Alternate hypoth
Thanks Robert for the review.
On Thu, Mar 16, 2017 at 8:05 PM, Robert Haas wrote:
> On Thu, Mar 16, 2017 at 8:28 AM, Rahila Syed
> wrote:
> > Thank you for the updated patch.
> >
> > I have applied and tested it on latest sources and the patch looks good
> to
> > me.
>
> The documentation puts
On 3/17/17 4:18 AM, Tsunakawa, Takayuki wrote:
> From: pgsql-hackers-ow...@postgresql.org
>> [mailto:pgsql-hackers-ow...@postgresql.org] On Behalf Of Tsunakawa,
>> Takayuki
>> I made this ready for committer. The patch applied except for catversion.h,
>> the patch content looks good, and the targe
Fabien COELHO writes:
> ISTM that I've tried to suggest to work around that complexity by:
> - document that \if-related commands should only occur at line start
> (and extend to eol).
> - detect and complain when this is not the case.
I think this is a lousy definition, and would never b
> Thank you for pointing out.
> Yeah, I agree that the twophase regression test should not be
> performed by default as long as the default value of
> max_prepared_transactions is 0. Similar to this, the isolation check
> regression test does same thing. Attached patch removes sql/twophase
> from s
On 03/08/2017 08:29 PM, Andrew Dunstan wrote:
>
> On 03/08/2017 08:33 AM, Peter Eisentraut wrote:
>> On 3/6/17 09:38, Peter Eisentraut wrote:
>>> On 3/4/17 01:45, Petr Jelinek wrote:
If that's the case, the attached should fix it, but I have no way of
testing it on windows, I can only s
Hi All,
I created a "Read-only" User in PostgreSQL via a Role with "SELECT" ONLY
privilege on all tables in a schema as shown below:
GRANT SELECT ON ALL TABLES IN SCHEMA [schema_name] TO [role_name]
GRANT [role_name] TO [user_name]
Next, I test this by trying to UPDATE a column in a table (same
I have committed the page_checksum function, will work on the bt stuff next.
I left in the superuser check, because I was not confident how well
pg_checksum_page() would handle messed up data.
--
Peter Eisentraut http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remo
On Fri, Mar 17, 2017 at 6:13 PM, Amit Kapila wrote:
> On Fri, Mar 17, 2017 at 12:27 PM, Ashutosh Sharma
> wrote:
>> On Fri, Mar 17, 2017 at 8:20 AM, Amit Kapila wrote:
>>
>> As I said in my previous e-mail, I think you need
>>> to record clearing of this flag in WAL record XLOG_HASH_DELETE as y
Peter Eisentraut wrote:
> On 3/16/17 11:56, Alvaro Herrera wrote:
> > Michael Paquier wrote:
> >
> >> What are you using as CFLAGS? As both typenames should be normally
> >> set, what about initializing those fields with NULL and add an
> >> assertion like the attached?
> >
> > Actually, my compi
Andrew Dunstan writes:
> I have confirmed on jacana Petr's observation that adding a timeout to
> the WaitLatchOrSocket cures the problem.
Does anyone have a theory as to why that cures the problem?
What length of timeout is being suggested here? Would a long timeout
perhaps create a performanc
Hi Osahon,
2017-03-17 15:54 GMT+01:00 Osahon Oduware :
> Hi All,
>
> I created a "Read-only" User in PostgreSQL via a Role with "SELECT" ONLY
> privilege on all tables in a schema as shown below:
>
> GRANT SELECT ON ALL TABLES IN SCHEMA [schema_name] TO [role_name]
> GRANT [role_name] TO [user_na
A handful of rather surprising errors showed up in our log extract this
morning, along the lines of:
2017-03-17 05:01:55 CDT [5400]: [1-1] @ FATAL: 57P01: terminating
connection due to administrator command
After a moment of more than a little astonishment, a look at the full log
revealed
Hello Tom,
ISTM that I've tried to suggest to work around that complexity by:
- document that \if-related commands should only occur at line start
(and extend to eol).
- detect and complain when this is not the case.
I think this is a lousy definition, and would never be considered if
Andres Freund writes:
> [ latest patches ]
I looked through 0001 (the composite-type-dependencies one). Although
I agree that it'd be good to tighten things up in that area, I do not
think we want this specific patch: it tightens things too much. Consider
this variant of the existing test case
>
>
> command.c:38:25: fatal error: conditional.h: No such file or directory
> #include "conditional.h"
>
Odd, it's listed as a new file in git status. Anyway, my point of posting
the WIP patch was to give people a reference point and spark discussion
about the next step, and it succeeded at that
Fabien COELHO writes:
>> I also fear that there are corner cases where the behavior would still
>> be inconsistent. Consider
>>
>> \if ...
>> \set foo `echo \endif should not appear here`
> In this instance, ISTM that there is no problem. On "\if true", set is
> executed, all is well. On "\if
Hi Giuseppe,
Thanks for the response. I have provided the GRANTS and other PostgreSQL
setup scripts below as it includes what you have suggested:
ROLE
-
CREATE ROLE WITH NOLOGIN NOSUPERUSER INHERIT NOCREATEDB
NOCREATEROLE NOREPLICATION;
USER
--
CREATE USER WITH PASSWORD ''
REV
On 17/03/17 16:07, Tom Lane wrote:
> Andrew Dunstan writes:
>> I have confirmed on jacana Petr's observation that adding a timeout to
>> the WaitLatchOrSocket cures the problem.
>
> Does anyone have a theory as to why that cures the problem?
>
I now have theory and even PoC patch for it.
The l
>
>
> I am sending the review of this patch
I found the following
v Use tage in documentation
v Don’t modified existing test case add new one instead
v Comment in pg_constraint.c is extended make it short
v Error message can be more guider if it tells about general rule
v Wrong result i
On Thu, Mar 16, 2017 at 10:34 PM, Craig Ringer wrote:
> On 17 March 2017 at 08:10, Stas Kelvich wrote:
>> While working on this i’ve spotted quite a nasty corner case with aborted
>> prepared
>> transaction. I have some not that great ideas how to fix it, but maybe i
>> blurred my
>> view and m
On Fri, Mar 17, 2017 at 11:42 AM, Tom Lane wrote:
> Fabien COELHO writes:
> >> I also fear that there are corner cases where the behavior would still
> >> be inconsistent. Consider
> >>
> >> \if ...
> >> \set foo `echo \endif should not appear here`
>
> > In this instance, ISTM that there is no
While I do not know QGIS, I'm wondering if it's similar to some of our
applications where they always use the same system login for the database
while each user provides a unique login to the application. Have you ever
set log_connections in your postgresql.conf file? That would show you which
user
Hi all,
2017-03-17 17:09 GMT+01:00 John Scalia :
> While I do not know QGIS, I'm wondering if it's similar to some of our
> applications where they always use the same system login for the database
> while each user provides a unique login to the application. Have you ever
> set log_connections i
On Fri, Mar 17, 2017 at 1:54 AM, Thomas Munro
wrote:
> SELECT *
>FROM lineitem l1
> WHERE EXISTS (SELECT *
> FROM lineitem l2
> WHERE l1.l_orderkey = l2.l_orderkey);
>
> -> estimates 59986012 rows, actual rows 59,986,052 (scale 10 TPCH)
>
> SELECT *
>
Hi John,
Thanks for your response. From my experience as a Software Developer, I
don't think QGIS can logon to my database/schema/table with a dedicated
user as they need authentication to do so. What you described about using a
dedicated user in applications is only possible because that user was
I'm struggling to find a good way to share code between
bt_page_items(text, int4) and bt_page_items(bytea).
If we do it via the SQL route, as I had suggested, it makes the
extension non-relocatable, and it will also create a bit of a mess
during upgrades.
If doing it in C, it will be a bit tricky
Petr Jelinek writes:
> Now the documentation for WSAEventSelect says "The FD_WRITE network
> event is handled slightly differently. An FD_WRITE network event is
> recorded when a socket is first connected with a call to the connect,
> ConnectEx, WSAConnect, WSAConnectByList, or WSAConnectByName fu
On 3/17/17 11:06, Alvaro Herrera wrote:
> I don't get anything with my compiler (gcc (Debian 4.9.2-10) 4.9.2) and
> I don't see anything in the few logs I looked at (clang, 9.6) from
> buildfarm either. What are you seeing specifically?
cc -Wall -Wmissing-prototypes -Wpointer-arith
-Wdeclaration-
Hi,
On 2017-03-17 11:36:30 -0400, Tom Lane wrote:
> Having said all that, I think that 0001 is contributing very little to the
> goals of this patch set. Andres stated that he wanted it so as to drop
> some of the one-time checks that execQual.c currently does for Vars, but
> I'm not really convi
Corey Huinker writes:
> I think Fabien was arguing that inside a false block there would be no
> syntax rules beyond "is the first non-space character on this line a '\'
> and if so is it followed with a if/elif/else/endif?". If the answer is no,
> skip the line. To me that seems somewhat similar
On 17/03/17 17:28, Tom Lane wrote:
> Petr Jelinek writes:
>> Now the documentation for WSAEventSelect says "The FD_WRITE network
>> event is handled slightly differently. An FD_WRITE network event is
>> recorded when a socket is first connected with a call to the connect,
>> ConnectEx, WSAConnect,
Hello Tom,
I also fear that there are corner cases where the behavior would still
be inconsistent. Consider
\if ...
\set foo `echo \endif should not appear here`
In this instance, ISTM that there is no problem. On "\if true", set is
executed, all is well. On "\if false", the whole line wou
>
>
> In the end, I suspect that teaching all the backslash commands to do
> nothing after absorbing their arguments is likely to be the least messy
> way to tackle this, even if it makes for a rather bulky patch.
>
>
Perhaps, but just glancing at \connect makes me think that for some
commands (pre
Robert Haas writes:
> The relevant code is in neqsel(). It estimates the fraction of rows
> that will be equal, and then does 1 - that number. Evidently, the
> query planner thinks that l1.l_suppkey = l2.l_suppkey would almost
> always be true, and therefore l1.l_suppkey <> l2.l_suppkey will alm
Hi,
Am Freitag, den 17.03.2017, 20:46 +0900 schrieb Michael Paquier:
> On Fri, Mar 17, 2017 at 7:18 PM, Michael Banck
> wrote:
> > Am Freitag, den 17.03.2017, 10:50 +0900 schrieb Michael Paquier:
> >> The comment block format is incorrect. I would think as well that this
> >> comment should say i
I wrote:
> The problem here appears to be that we don't have any MCV list for
> the "twothousand" column (because it has a perfectly flat distribution),
> and the heuristic that eqjoinsel_semi is using for the no-MCVs case
> is falling down badly.
Oh ... wait. eqjoinsel_semi's charter is to "esti
On 17 March 2017 at 01:37, Robert Haas wrote:
> - You've added a GUC (which is good) but not documented it (which is
> bad) or added it to postgresql.conf.sample (also bad).
>
> - You've used a loop inside a spinlock-protected critical section,
> which is against project policy. Use an LWLock; de
On 16 March 2017 at 18:18, Ashutosh Bapat
wrote:
> + * Check if we are already finished plans from parallel append. This
> + * can happen if all the subplans are finished when this worker
> + * has not even started returning tuples.
> + */
> +if (node->as_pa
Andres Freund writes:
> That said, it seems this is something that has to wait for a later
> release, I'm putting back in similar logic as there was before (not a
> branch, but change the opcode to a non-checking variant).
Yeah, I was wondering if changing the opcode would be preferable to
a firs
I gave this patch a quick skim. At first I was confused by the term
"catalog"; I thought it meant we stored options in a system table. But
that's not what is meant at all; instead, what we do is build these
"catalogs" in memory. Maybe a different term could have been used, but
I'm not sure it's
Petr Jelinek writes:
> On 17/03/17 17:28, Tom Lane wrote:
>> Yeah, I'm afraid we had better do something more or less like this.
>> It's interesting to speculate about whether WaitEventSet could keep
>> additional state that would avoid the need for a dummy send() every
>> time, but that seems lik
While trying to figure out some bloating in the newly logged hash indexes,
I'm looking into the type of each page in the index. But I get an error:
psql -p 9876 -c "select hash_page_type(get_raw_page('foo_index_idx',x))
from generate_series(1650,1650) f(x)"
ERROR: page is not a hash page
DETAIL
Corey Huinker writes:
>> In the end, I suspect that teaching all the backslash commands to do
>> nothing after absorbing their arguments is likely to be the least messy
>> way to tackle this, even if it makes for a rather bulky patch.
> Perhaps, but just glancing at \connect makes me think that f
While reviewing Ashutosh Bapat's partitionwise join code, I noticed
he'd run up against the problem that adjust_relid_set() is defined as
static in two different source files, and he wanted to call it from a
third file. I didn't much like his solution to that problem, which
was to rename one of th
Peter Eisentraut wrote:
> cc -Wall -Wmissing-prototypes -Wpointer-arith
> -Wdeclaration-after-statement -Wendif-labels -Wmissing-format-attribute
> -Wformat-security -fno-strict-aliasing -fwrapv
> -fexcess-precision=standard -g -O2 -I../../../src/include
> -D_FORTIFY_SOURCE=2 -DLINUX_OOM_ADJ -D_GN
Robert Haas writes:
> While reviewing Ashutosh Bapat's partitionwise join code, I noticed
> he'd run up against the problem that adjust_relid_set() is defined as
> static in two different source files, and he wanted to call it from a
> third file. I didn't much like his solution to that problem,
Hi,
Currently, clients wishing to know when the server exits hot standby
have to resort to polling, which is often suboptimal.
This adds the new "in_hot_standby" GUC variable that is reported via a
ParameterStatus message. This allows the clients to:
(a) know right away that they are connect
Heikki Linnakangas writes:
> Fix and simplify check for whether we're running as Windows service.
This seems to have broken narwhal:
Creating library file: libpostgres.a
../../src/port/libpgport_srv.a(win32security_srv.o)(.text+0x109): In function
`pgwin32_is_admin':
C:/msys/1.0/local/pgbuildfa
On Fri, Mar 17, 2017 at 1:50 PM, Tom Lane wrote:
> Robert Haas writes:
>> While reviewing Ashutosh Bapat's partitionwise join code, I noticed
>> he'd run up against the problem that adjust_relid_set() is defined as
>> static in two different source files, and he wanted to call it from a
>> third
>
>
> > \set x 'arg1 arg2'
>
> > \if false
> > \cmd_that_takes_exactly_two_args :x
> > \endif
>
> Yeah, throwing errors for bad arguments would also need to be suppressed.
>
> regards, tom lane
>
Ok, barring other feedback, I'm going to take my marching orders as "make
On Fri, Mar 17, 2017 at 1:14 PM, Tom Lane wrote:
> After a bit more thought, it seems like the bug here is that "the
> fraction of the LHS that has a non-matching row" is not one minus
> "the fraction of the LHS that has a matching row". In fact, in
> this example, *all* LHS rows have both matchi
With some intensive crash-recovery testing, I've run into a situation where
I get some bad table bloat. There will be large swaths of the table which
are empty (all results from heap_page_items other than lp are either zero
or NULL), but have zero available space in the fsm, and are marked as
all-
On Fri, Mar 17, 2017 at 10:54 PM, Jeff Janes wrote:
> While trying to figure out some bloating in the newly logged hash indexes,
> I'm looking into the type of each page in the index. But I get an error:
>
> psql -p 9876 -c "select hash_page_type(get_raw_page('foo_index_idx',x)) from
> generate_s
1 - 100 of 133 matches
Mail list logo