amount of
types)
and an alignment of the type (which is currently not really needed since
the abi specifies that alignof == sizeof for all scalar types). */
Specifically "the abi specifies that alignof == sizeof for all scalar types"
Vladimir Kokovic
Belgrade, 11.Nov
bloat is not
solved.
That is what I mean.
Vladimir
de a way to do so.
Adding protocol messages would blow pgbouncer, etc things, so it makes
sense to refrain from new messages unless it is absolutely required.
Vladimir
>OK, so here's a patch. Review appreciated.
Please correct typo "Write an unsigned integer field (anythign written with
UINT64_FORMAT)". anythign -> anything.
Vladimir
ons, so
it would be easier to reson about
Have you tried "withStatusInterval(20, TimeUnit.SECONDS)" instead of 20
millis? I don't think it matter much, however 20ms seems to be an overkill.
Vladimir
пт, 15 сент. 2017 г. в 19:57, Dipesh Dangol :
> hi,
>
> I am trying to im
> 5 сент. 2017 г., в 15:48, Michael Paquier
> написал(а):
>
> On Tue, Sep 5, 2017 at 9:40 PM, Vladimir Borodin wrote:
>> We do compress WALs and send them over network. Doing it via archive_command
>> in single thread is sometimes slower than new WALs are written unde
> 5 сент. 2017 г., в 15:42, Chris Travers написал(а):
>
> On Tue, Sep 5, 2017 at 2:40 PM, Vladimir Borodin <mailto:r...@simply.name>> wrote:
>
>> 5 сент. 2017 г., в 14:04, Michael Paquier > <mailto:michael.paqu...@gmail.com>> написал(а):
>>
>&g
> 5 сент. 2017 г., в 14:04, Michael Paquier
> написал(а):
>
>> For example, in archive_command we put WALs for archiving from
>> pg_xlog/pg_wal into another directory inside PGDATA and than another cron
>> task makes real archiving. This directory ideally should be skipped by
>> pg_rewind, but
> 5 сент. 2017 г., в 12:31, Chris Travers написал(а):
>
> I think the simplest solution for now is to skip any files ending in .conf,
> .log, and serverlog.
Why don’t you want to solve the problem once? It is a bit harder to get
consensus on a way how to do it, but it seems that there are no
is caught by checksums. I am
not familiar with WAL protocol - do we have enough checksums when writing
it out and on the wire? I suspect there are much more things PostgreSQL can
do to be more resilient, and at least detect corruptions earlier.
--
Vladimir Rusinov
PostgreSQL SRE, Google Ireland
Goog
Hi, Chris.
> 10 авг. 2017 г., в 15:09, Chris Travers написал(а):
>
> Hi;
>
> I ran into a funny situation today regarding PostgreSQL replication and wal
> corruption and wanted to go over what I think happened and what I wonder
> about as a possible solution.
>
> Basic information is custom-
I haven't seen that but can't guarantee that such case does not exist
28 июля 2017 г. 9:19 PM пользователь "Alvaro Herrera" <
alvhe...@2ndquadrant.com> написал:
> Vladimir Kunschikov wrote:
> > >This "maxlen" business and the fallback error messa
6df51cf17c0b0a4f201ee0f93cf258de9#diff-98e3f8ce5d6e87950dd66e4c8bdedb21R713
It was rewritten for the sake of somewhat exaggerated security.
Version #5 in attachment.
--
Regards,
Vladimir Kunschikov
diff --git a/src/bin/pg_dump/compress_io.c b/src/bin/pg_dump/compress_io.c
index 991fe11..b8af1bc 100644
--- a/src/b
Hello Alvaro,
here goes v4 version: removed unused header.
Compilation of this code snippet with -Wall -Wexter -std=c89 doesn't produce
any warnings.
--
Best regards,
Vladimir Kunschikov
Lead software developer
IDS project
InfoTeCS JSC
From: Kunshc
s.int/company/structure/Pages/detailsDepartment.aspx?idDepartment=235&idCompany=17>
From: Alvaro Herrera
Sent: Wednesday, July 26, 2017 7:40:20 PM
To: Kunshchikov Vladimir
Cc: pgsql-hackers@postgresql.org
Subject: Re: [HACKERS] [patch] pg_dump/pg_restore zerror() a
Hello Alvaro, thanks for the feedback, fixed all of your points.
Attached new version of patch.
--
Best regards,
Vladimir Kunschikov
Lead software developer
IDS project
InfoTeCS JSC
From: Alvaro Herrera
Sent: Wednesday, July 26, 2017 1:02 AM
To
> 14 июля 2017 г., в 1:33, Stephen Frost написал(а):
>
> What would be really nice for such cases is support for Kerberos and
> delegated Kerberos credentials. Having pgpool support that would remove
> the need to deal with passwords at all.
Since nearly all systems with some kind of load nowa
input file: d3/2811.dat.gz:
invalid distance too far back
Attached small fix for this issue.
You can view that patch online on our github:
https://github.com/Infotecs/postgres/commit/1578f5011ad22d78ae059a4ef0924426fd6db762
--
Best regards,
Vladimir Kunschikov
Lead software developer
I
BEGIN:VCALENDAR
PRODID:-//Google Inc//Google Calendar 70.9054//EN
VERSION:2.0
CALSCALE:GREGORIAN
METHOD:REQUEST
BEGIN:VEVENT
DTSTART:20170616T06Z
DTEND:20170616T07Z
DTSTAMP:20170615T193848Z
ORGANIZER;CN=Vladimir Sitnikov:mailto:sitnikov.vladi...@gmail.com
UID:2a6cc082-5202-11e7-978f
> 12 июня 2017 г., в 13:19, Amit Kapila написал(а):
>
> On Sun, Jun 11, 2017 at 11:59 PM, Vladimir Borodin <mailto:r...@simply.name>> wrote:
>>
>> 8 июня 2017 г., в 17:03, Amit Kapila написал(а):
>>
>> On Thu, Jun 8, 2017 at 6:49 PM, Dmitriy Sar
> 8 июня 2017 г., в 17:03, Amit Kapila написал(а):
>
> On Thu, Jun 8, 2017 at 6:49 PM, Dmitriy Sarafannikov
> wrote:
>>
>>> Why didn't rsync made the copies on master and replica same?
>>
>> Because rsync was running with —size-only flag.
>>
>
> IIUC the situation, the new WAL and updated p
> 6 июня 2017 г., в 23:30, Sergey Burladyan написал(а):
>
> Dmitriy Sarafannikov writes:
>
>> Starting and stopping master after running pg_upgrade but before rsync to
>> collect statistics
>> was a bad idea.
>
> But, starting and stopping master after running pg_upgrade is *required*
> by d
> 27 мая 2017 г., в 19:56, Andres Freund написал(а):
>
> On 2017-05-27 19:48:24 +0300, Vladimir Borodin wrote:
>> Well, actually clean shutdown of master with exit code 0 from `pg_ctl
>> stop -m fast` guarantees that all WAL has been replicated to standby.
>
> It do
> 26 мая 2017 г., в 21:39, Amit Kapila написал(а):
>
>> And LSN on replica is greater that LSN on master (838D/C4A0D280 >
>> 8092/6A26DD08)
>> How can this be possible?
>>
>
> Yeah, I think this is quite suspicious. This seems to indicate that
> not all WAL records are replicated before the
> 26 мая 2017 г., в 23:04, Michael Paquier
> написал(а):
>
> On Fri, May 26, 2017 at 2:39 PM, Amit Kapila wrote:
>> Yeah, I think this is quite suspicious. This seems to indicate that
>> not all WAL records are replicated before the switchover. What is the
>> value of "synchronous_commit" yo
> 24 мая 2017 г., в 15:44, Robert Haas написал(а):
>
> On Wed, May 24, 2017 at 7:27 AM, Dmitriy Sarafannikov
> wrote:
>> It seems like replica did not replayed corresponding WAL records.
>> Any thoughts?
>
> heap_xlog_freeze_page() is a pretty simple function. It's not
> impossible that it co
Hi.
> 25 апр. 2017 г., в 18:13, Dmitriy Sarafannikov
> написал(а):
>
> I'd like to propose to search min and max value in index with SnapshotAny in
> get_actual_variable_range function.
> Current implementation scans index with SnapshotDirty which accepts
> uncommitted rows and rejects dead r
> 29 апр. 2017 г., в 17:34, Tom Lane написал(а):
>
> Dmitriy Sarafannikov writes:
>>> Maybe we need another type of snapshot that would accept any
>>> non-vacuumable tuple. I really don't want SnapshotAny semantics here,
>
>> If I understood correctly, this new type of snapshot would help if
Please disregard my previous message.
pgjdbc is already doing upcase conversion, so I would like to see a test
case that reproduces the error.
Alternatively, could you please capture and share TRACE log? (
https://jdbc.postgresql.org/documentation/head/logging.html#configuration )
Vladimir
ср
legrand>when usingversion 42.0.0 with
legrand> jdbc:postgresql://localhost:5432/postgres?autosave=always
The pitfall there is the value should be written with upper case like
autosave=ALWAYS.
I've filed https://github.com/pgjdbc/pgjdbc/issues/769 to improve that at
some point.
Vladimir
and human-
readable and writable and generally easy to debug. I'm not bold enough to
propose PostgreSQL to use it, but I'd be happy to see ideas from it to be
used elsewhere.
--
Vladimir Rusinov
Storage SRE, Google Ireland
Google Ireland Ltd.,Gordon House, Barrow Street, Dublin 4, Ireland
Registered in Dublin, Ireland
Registration Number: 368047
smime.p7s
Description: S/MIME Cryptographic Signature
> 6 февр. 2017 г., в 4:57, Peter Geoghegan написал(а):
>
> I meant that I find the fact that there were no user reports in all
> these years to be a good reason to not proceed for now in this
> instance.
Well, we had some strange situations with indexes (see below for example) but I
couldn’t e
> 31 янв. 2017 г., в 9:50, Michael Paquier
> написал(а):
>
> On Mon, Jan 30, 2017 at 4:28 PM, Andrew Borodin wrote:
>> I'll summarize here my state of studying concurrent methods of page
>> unlinking.
>>
>> GIN B-tree does not have "high key". That means, that rightmost key on
>> a page is m
m now
given there are patches above that rename other things as well.
--
Vladimir Rusinov
Storage SRE, Google Ireland
Google Ireland Ltd.,Gordon House, Barrow Street, Dublin 4, Ireland
Registered in Dublin, Ireland
Registration Number: 368047
smime.p7s
Description: S/MIME Cryptographic Signature
all uses of "xlog" to
> > "wal". What about program names, command line options, etc.? If the
> > argument is, we changed one thing, we should change the rest, then let's
> > see that. I think that argument itself is flawed, but if that's what
> >
> 21 янв. 2017 г., в 18:18, Petr Jelinek
> написал(а):
>
> On 21/01/17 11:39, Magnus Hagander wrote:
>> Is it time to enable checksums by default, and give initdb a switch to
>> turn it off instead?
+1
>
> I'd like to see benchmark first, both in terms of CPU and in terms of
> produced WAL (
Core dump was extracted from this build
https://travis-ci.org/Gordiychuk/pgjdbc/builds/193318018
2017-01-19 17:38 GMT+03:00 Vladimir Gordiychuk :
> Last success build from postgresql HEAD was 20 days ago
> https://travis-ci.org/pgjdbc/pgjdbc/builds/187764274
> We detect this problem 3
Last success build from postgresql HEAD was 20 days ago
https://travis-ci.org/pgjdbc/pgjdbc/builds/187764274
We detect this problem 3 days ago.
2017-01-19 17:22 GMT+03:00 Dave Cramer :
> The travis job https://travis-ci.org/pgjdbc/pgjdbc/jobs/192517342 is
> seeing a segfault when we are testing a
On Wed, Jan 18, 2017 at 12:28 PM, Michael Paquier wrote:
> On Wed, Jan 18, 2017 at 8:15 PM, Vladimir Rusinov
> wrote:
> > On the topic of binaries, there's going to be another patch renaming
> them.
> > Those will have no aliases as it's trivial to work-around
k-around (symlinks, shell
scripts, etc) and not so trivial to implement in a portable way.
I'm used to workflows with lots of small commits, so I was not able to
force myself to include all of the renames in one diff.
--
Vladimir Rusinov
Storage SRE, Google Ireland
Google Ireland Ltd.,Gordon
e maintenance burden than just keeping aliases.
--
Vladimir Rusinov
Storage SRE, Google Ireland
Google Ireland Ltd.,Gordon House, Barrow Street, Dublin 4, Ireland
Registered in Dublin, Ireland
Registration Number: 368047
smime.p7s
Description: S/MIME Cryptographic Signature
e the world: Stephen, Tom, David F, Vik, Bruce M => 5
> - Yes to aliases: Michael P, Andres, Peter E., Cynthia S, Jim N,
> Vladimir, Simon R, Fujii-san => 8
> If I misunderstood things, please feel free to speak up.
>
Counting amendments and more votes down the thread:
- no aliases:
ts on this thread, keeping your patch
> with the compatibility table is the best move for now. Even if we end
> up by having a version without aliases, that will be just code to
> remove in the final version.
>
Indeed, it is trivial to kill aliases.
New version of the patch attached.
e
pg_xlogfile_name| pg_wal_file_name
pg_xlogfile_name_offset | pg_wal_file_name_offset
--
Vladimir Rusinov
Storage SRE, Google Ireland
Google Ireland Ltd.,Gordon House, Barrow Street, Dublin 4, Ireland
Registered in Dublin, Ireland
Registration Number: 368047
diff --git a/contrib/bloom
used
pg_last_xlog_receive_location | pg_last_wal_receive_location
pg_last_xlog_replay_location| pg_last_wal_replay_location
pg_switch_xlog | pg_switch_wal
pg_xlog_location_diff | pg_wal_location_diff
pg_xlog_replay_pause | pg_wal_replay_pause
pg_xlog_replay_
function names. I've used 'recovery' instead of 'xlog_replay' and
used 'wal_file' instead of 'xlogfile'. Does it make sense?
- Release notes. I was unable to find a draft for 10.0. How do I make sure
these renames are not forgotten?
--
Vladimir Rusinov
Storag
rd
deadline for removal in 11.0.
This way we are nicer to people who maintain their tools and read release
notes via giving them more time, and nicer to ourselves via cleaning legacy
stuff relatively soon.
--
Vladimir Rusinov
Storage SRE, Google Ireland
Google Ireland Ltd.,Gordon House, Barro
nd maybe other hosting providers)
and ask them to migrate to new function names. This is of course will not
fix everything, but it would be a significant chunk. I suspect a lot of
private tools/scripts borrow snippets from github as well.
--
Vladimir Rusinov
Storage SRE, Google Ireland
Go
On Thu, Dec 29, 2016 at 5:59 PM, Stephen Frost wrote:
> I have a pretty strong
> feeling about this particular change being worthwhile and also pretty
> long overdue.
>
Yeah, sorry for that. I should be able to make some progress early January.
--
Vladimir Rusinov
Storage SRE, Go
Now, I'm not sure whether it is worth maintaining function aliases.
Assuming these are indeed trivial (can somebody point me to example?) I see
roughly the same amount of downsides both ways.
Having aliases raises additional questions:
- do we keep them documented (probably not?)
- do we keep th
ests pass.
I have no further comments, although since I'm not experienced Postgres
reviewer, commiter may want to take another look before merging.
--
Vladimir Rusinov
Storage SRE, Google Ireland
Google Ireland Ltd.,Gordon House, Barrow Street, Dublin 4, Ireland
Registered in Dublin, Irelan
es little sense.
Ack. I was not sure what patch size is preferred here. Will continue with a
patch in original thread later.
Thanks for feedback!
--
Vladimir Rusinov
Storage SRE, Google Ireland
Google Ireland Ltd.,Gordon House, Barrow Street, Dublin 4, Ireland
Registered in Dublin, Ir
the need to consult the log archive, thus
+making this a completely standalone backup.
I suggest "method none" instead of "option
none". I found the word "option" confusing in that
sentence.
--
Vladimir Rusinov
Storage SRE, Google Ireland
G
cally
freezes behavior of psql and makes e.g. server tests depend not only server
behavior but also on piece of irrelevant client-only code.
I could imagine a test system that is both has more-or-less human-readable
expected.out files and does not depend on exact decorations added by psql.
--
Vlad
rge just these, as it will not help
people with whitespace-eating editors.
--
Vladimir Rusinov
Storage SRE, Google Ireland
Google Ireland Ltd.,Gordon House, Barrow Street, Dublin 4, Ireland
Registered in Dublin, Ireland
Registration Number: 368047
smime.p7s
Description: S/MIME Cryptographic Signature
output files as a
separate issue.
One way to ease the pain is to remove trailing whitespaces in all supported
branches via separate patches.
--
Vladimir Rusinov
Storage SRE, Google Ireland
Google Ireland Ltd.,Gordon House, Barrow Street, Dublin 4, Ireland
Registered in Dublin, Ireland
Registrati
On Wed, Dec 14, 2016 at 4:50 PM, Tom Lane wrote:
> Vladimir Rusinov writes:
> > Therefore, I propose this patch.
>
> Right now is a really bad time to do that; what it will mostly accomplish
> is to break back-patching of doc fixes for little benefit.
>
ack. As I said, it
They are considered bad practice in many style guides and many editors
configured to stip them on every save.
Such editors will produce spurious diffs when editing the documentation.
Therefore, I propose this patch.
--
Vladimir Rusinov
Storage SRE, Google Ireland
Google Ireland Ltd.,Gordon
On Wed, Dec 14, 2016 at 4:07 PM, Peter Eisentraut <
peter.eisentr...@2ndquadrant.com> wrote:
> On 12/13/16 12:47 PM, Vladimir Rusinov wrote:
> > Based on discussion in
> > https://www.postgresql.org/message-id/CAE1wr-w%
> 3DLE1cK5uG_rmAh-VBxc4_Bnw-gAE3qSqL-%3Dt
Based on discussion in
https://www.postgresql.org/message-id/CAE1wr-w%3DLE1cK5uG_rmAh-VBxc4_Bnw-gAE3qSqL-%3DtWwvLvjQ%40mail.gmail.com
.
Tested via regression tests.
To be applied in master only and to be included in 10.0.
--
Vladimir Rusinov
Storage SRE, Google Ireland
Google Ireland Ltd
> 6 дек. 2016 г., в 19:19, David Fetter написал(а):
>
> On Tue, Dec 06, 2016 at 10:52:47AM -0500, Tom Lane wrote:
>> Vladimir Borodin writes:
>>> Any chance to get tzdata 2016j in supported branches?
>>
>> When the next scheduled releases come around (Fe
Hi all.
Any chance to get tzdata 2016j in supported branches?
--
May the force be with you…
https://simply.name
On Fri, Dec 2, 2016 at 1:59 PM Michael Paquier
wrote:
> On Fri, Dec 2, 2016 at 8:28 PM, Vladimir Rusinov
> wrote:
> > I guess it would make sense to do all of it in 10.0.
> > I'm new here, so not very sure about process. How many commit fests
> could I
> > expe
On Fri, Dec 2, 2016 at 2:02 AM Michael Paquier
wrote:
> On Fri, Dec 2, 2016 at 3:21 AM, Robert Haas wrote:
> > On Thu, Dec 1, 2016 at 10:29 AM, Vladimir Rusinov
> wrote:
> >> I've found myself wondering "where is my xlog" after running
> >> pg_swit
_recievexlog to pg_revievewal (and others in bin/)?
- rename pg_xlogdump to pg_waldump?
- if we do rename, should we keep aliases for functions and symlinks for
tools?
- anything else?
--
Vladimir Rusinov
Storage SRE, Google Ireland
Google Ireland Ltd.,Gordon House, Barrow Street, Dublin 4, Irelan
Performance review
===
n/a
Coding review
===
LGTM
Architecture review
==
n/a
--
Vladimir Rusinov
Bigtable SRE
Google Ireland Ltd.,Gordon House, Barrow Street, Dublin 4, Ireland
Registered in Dublin, Ireland
Registration Number: 368047
On Mon, Oct 3, 2016 at 5:44 AM, Micha
> 28 нояб. 2016 г., в 20:31, Andrew Borodin написал(а):
>
> This patch solved a problem encountered by Evgeniy Efimkin and
> Vladimir Borodin from Yandex.Mail.
>
> and eventually deleting some of data. This testbed showed VACUUM
> holding inserts for up to tenths of sec
:
> On Mon, Nov 21, 2016 at 6:32 AM, Vladimir Svedov
> wrote:
>
>> Hi,
>> I have this question. Looked for a help on http://dba.stackexchange.com/
>> No success.
>> Maybe you can answer?..
>> Thank you in advance
>>
>>
>> "FOREIGN
No, I select from it OK.
The bug(?) is that when I do it in oppened psql session it produces no log,
and when I run same select as psql -c "SELECT..." it gives the above
2016-11-22 20:18 GMT+00:00 Robert Haas :
> On Tue, Nov 22, 2016 at 5:05 AM, Vladimir Svedov
> wrote:
&
Hi,
Sorry - tried to reproduce on other machine and gather all statements. And
failed
Installed 9.3 (which has those symptoms) and still can't reproduce.
Must be platform specific, not version
2016-11-21 21:58 GMT+00:00 Kevin Grittner :
> On Mon, Nov 21, 2016 at 8:32 AM, Vladimir Svedov
Hi,
I have this question. Looked for a help on http://dba.stackexchange.com/
No success.
Maybe you can answer?..
Thank you in advance
"FOREIGN_TABLE" created with postgres_fdw. LOCAL_TABLE is just a local
table...
Symptoms:
1. I run in psql query SELECT * from FOREIGN_TABLE. No log generated
>>> When sending a big message, WalSndWriteData() notices that it's
>>> approaching timeout and tries to send a keepalive request, but the
>>> request just gets buffered behind the remaining output plugin data and
>>> isn't seen by the client until the client has received the rest of the
>>> pendin
Craig, Andres what do you thinks about previous message?
2016-10-06 0:35 GMT+03:00 Vladimir Gordiychuk :
> > Vladimir? I'm running out of time to spend on this at least until the next
> CF. Think you can make these changes?
>
> Yes, I can. But I thinks It sho
>What platform does the postgres server run on?
Ubuntu
OS name: "linux", version: "3.19.0-66-generic", arch: "amd64", family:
"unix"
2016-10-18 11:05 GMT+03:00 Magnus Hagander :
>
>
> On Tue, Oct 18, 2016 at 1:00 AM, Vladimir Sitnikov <
org/pgjdbc/pgjdbc/builds/168444341
Any hints what could be the issue?
Was the breakage intentional?
Vladimir
> Vladimir? I'm running out of time to spend on this at least until the next
CF. Think you can make these changes?
Yes, I can. But I thinks It should be discuss first.
> Terminating COPY BOTH with ERRCODE_QUERY_CANCELED seems fine. If it
> was expecting the error the client c
> Vladimir, can I have your opinion on the latest patch or if you want to
make changes, an updated patch?
I am satisfied with all our changes and I thinks it enough to complete this
PR.
2016-10-03 6:51 GMT+03:00 Craig Ringer :
> On 3 Oct. 2016 10:15, "Michael Paquier" wrote
, before this changes stopping logical replication
gets N ms but after apply changes it gets NN ms where NN ms less than N ms.
Is it available add this kind of benchmark to postgresql? I will be grateful
for links.
2016-09-26 5:32 GMT+03:00 Craig Ringer :
> On 26 Sep. 2016 02:16, "Vladimir Go
celTime) <= 3, # allow extra time for slow machines
"pg_recvlogical exited promptly on sigint when idle"
);
```
Also I do not understand why we do
$proc->pump while $proc->pumpable;
after send signal to process, why we can not just wait stop replication
slot?
2016-09-
immediately without wait CopyDone from
server.
0004 patch contain regression tests on scenarios that fix 0001 and 0002
patches.
2016-09-16 10:11 GMT+03:00 Vladimir Gordiychuk :
> >Have you had a chance to look at adding the tests we discussed?
>
> Not yet. I plane do it on this weekends
>Have you had a chance to look at adding the tests we discussed?
Not yet. I plane do it on this weekends
2016-09-16 4:37 GMT+03:00 Craig Ringer :
> On 9 September 2016 at 12:03, Craig Ringer wrote:
>
> > Setting "waiting on author" in CF per discussion of the need for tests.
>
> Have you had a
9-08 12:29 GMT+03:00 Craig Ringer :
> On 7 September 2016 at 15:44, Vladimir Gordiychuk
> wrote:
>
> > Fixed patch in attach.
>
> It'd helpful if you summarize the changes made when posting revisions.
>
> >> * There are no tests. I don't see any really simple way
and assume that majority of the rows have length(vc)<128.
Under that conditions, it would be nice to do index-only-scan
and filter (like in my previous case), but detect "long" rows
and do additional recheck for them.
Vladimir
There are no tests. I don't see any really simple way to test this,
> though.
>
> I suggest that you apply these updated/split patches to a fresh copy
> of master then see if you can update it based on this feedback.
> Something like:
>
> git checkout master
> git pull
>
;.
As shown in "query 2" vs "query 3", PostgreSQL cannot come up with such a
plan on its own
for some reason.
This is not a theoretical issue, but it is something that I use a lot with
Oracle DB (it just creates a good plan for "query 2").
Vladimir
logical replication.
2016-05-11 7:25 GMT+03:00 Craig Ringer :
> On 11 May 2016 at 06:47, Vladimir Gordiychuk wrote:
>
>> Same thread, I just think these are two somewhat separate changes. One is
>>> just in the walsender and allows return to command mode during waiting for
>&
ter: 977
Buffers: shared hit=17
-> Index Only Scan using type_vc__vlsi on vlsi (cost=0.55..67.97
rows=971 width=36) (actual time=0.015..0.210 rows=1000 loops=1)
Index Cond: (type = '42'::numeric)
Heap Fetches: 0
Buffers: shared hit=17
Planning time: 0.086 ms
Execution time: 0.503 ms
Vladimir
postgresql.org/docs/9.5/static/protocol-flow.html#PROTOCOL-FLOW-EXT-QUERY
doc>If successfully created, a named portal object lasts till the end of
the current transaction, unless explicitly destroyed
Vladimir
first 5 executions of PreparedStatement use unnamed
statements (Parse/Bind/Exec).
Then pgjdbc assigns a name and uses just Bind/Exec.
So if a particular SQL is rare, then it would not get its own
server-prepared name
even though it is "java.sql.PreparedStatement".
What pgjdbc does is it picks the
Vladimir>> Yes, that is what happens.
Vladimir>> The idea is not to mess with gucs.
Shay:> Wow... That is... insane...
Someone might say that "programming languages that enable side-effects
are insane".
Lots of connection pools work by sharing the connections and it is
atements
should be server-prepared and which should not.
Vladimir>> My experience shows, that people are very bad at predicting where the
Vladimir>> performance problem would be.
Vladimir>> For 80% (or even 99%) of the cases, they just do not care
thinking if a
Vladimir>>
t requires some discipline like "don't mess with gucs", however
that is acceptable and it is easily understandable by developers.
My application cannot afford re-parsing hot SQL statements as hurts
performance. It is very visible in the end-to-end performance tests, so
"discard all" is not used, and developers know not to mess with gucs.
Vladimir
quot;.
It is expected, that "driver" would "optimize" the handling somehow in the
best possible way.
It is Java API specification that enables me (as a driver developer) to be
flexible, and leverage database features so end user gets best experience.
Vladimir
>
ges in
the future if "pre-connect client_encoding" would be implemented somehow.
Vladimir
SQL versions (e.g. include
encoding name right into the error message).
2) Implement workaround for current drivers, so clients would get proper
error messages even when trying to connect to unpatched server.
Vladimir
n case of cached query cannot change result type error the
backend would skip processing the subsequent bind/exec kind of messages for
the problematic statement, and it will keep the transaction in open state.
The client would have to re-send relevant prepare/bind/execute". That would
prevent user from receiving the error and that approach does not require
savepoints.
This particular topic "how to fix <> error"
might be better to be split into its own conversation.
Vladimir
vs encoding? Maybe we need a ground for the list.
I use "locale -a" for that.
For instance, for Japanese it prints the following on my machine (OS X
10.11.6):
locale -a | grep ja
ja_JP
ja_JP.eucJP
ja_JP.SJIS
ja_JP.UTF-8
[1]:
https://github.com/pgjdbc/pgjdbc/commit/ec5fb4f5a66b6598aea1c7ab8df3126ee77d15e2#diff-57ed15f90f50144391f1c134bf08a45cR47
Vladimir
driver
developers.
Do you mean in C# world major connection pools default to "discard all"
setup? That sounds strange to me.
Vladimir
>
" message.
I hope you would forgive me if I just stop the discussion here.
I find I'd better spent that time on just fixing pgbouncer issue rather
than discussing if it is pgbouncer's or postgresql's issue.
I'm sorry for being impolite if I was ever.
Vladimir
>
what insane to assume people would use naked SQL. Of course they
would use ORMs and alike, so they just would not be able to pass that
additional parameter telling if a particular query out of thousands should
be server-prepared or not.
Vladimir> "cached plan cannot change result type&quo
1 - 100 of 273 matches
Mail list logo
