e
TimeZone | Host | configuration file
work_mem | 162MB | configuration file
(46 rows)
Regards, Mikhail
" to 1000. That gave me
the value of estimated number of rows for subscr_param_pkey closer to the real
value (est.64, real 30) which affected the execution plan in the right way.
>Thursday, June 29, 2017 7:21 PM +03:00 from Mikhail :
>
>
>Hi guys,
>I
. And the
miss
in cost estimation is because of the estimation of the number of rows to
return (est. 923 - real 30).
And my question:
1. am i right in my hypothesis?
2. is there a way to affect the cost evaluation in my case to help postgresql
choose the right execution plan?
3. is there a way to fix this query and not to break the execution of other
queries?
ps: running "analyze" on both tables doesn't affect the result
pps: all "cost" parameters are in their default values: cpu_index_tuple_cost,
cpu_operator_cost, cpu_tuple_cost, random_page_cost, seq_page_cost.
Thanks, Mikhail
--
9.4.10 + fixes
9.4.10 --> 9.6.1
>Wednesday, December 14, 2016 12:55 PM +03:00 from Thomas Kellerer
>:
>
>Mikhail schrieb am 13.12.2016 um 10:57:
>> Should i check all the production environments for the problems,
>> mentioned in all interim versions release notes, is it enoug
John, thanks! Your approach significantly reduces the number of checks.
>Вторник, 13 декабря 2016, 13:34 +03:00 от John R Pierce :
>
>On 12/13/2016 1:57 AM, Mikhail wrote:
>> Should i check all the production environments for the problems,
>> mentioned in all interim versi
) or there is another quick way to check if
i should apply some additional actions to standard pg_upgrade+stats_collect?
Regards, Mikhail Balayan
also restore; 2. they restore with their
locks on database objects; 3. if you're restoring using PITR, you should
rollback all the prepared transactions after database start.
Thank you, Tom.
Regards, Mikhail.
>Вторник, 13 октября 2015, 9:29 -04:00 от Tom Lane :
>
>=?UTF-8?B?TW
Sorry for my previous message, it went to wrong thread. Creating this as a new
one.
I'm running supposedly an easy operation - renaming index and my session hangs
on it. pg_stat_activity shows that process is active and is waiting for lock:
=# select application_name, waiting, state, query from
d
top shows the same: no action performing by process.
The situation is reproducible: i can interrupt the process, run index renaming
again and i'll get the same situation, described above.
Is there any ideas what is happening inside the database?
Regards, Mikhail
Hi,
I have a long executing query via foreign data wrapper and on the remote server
i can see that the query is fetching data from the cursor
=# select application_name, state, query from pg_stat_activity where pid =
15455;
application_name │ state │ query
──┼┼─
gresql.core.v3.QueryExecutorImpl.execute(QueryExecutorImpl.java:257)
Could you please help me to figure out the reason of such a failure. (This
bug happens once per 10 successful cases.)
Thanks in advance!
Best regards, Mike Pryakhin.
--
Regards, Mikhail Pryakhin.
gint.
Is there any reasonable way to do it in sql/pgplsql?
Thanks in advance.
--
Best regards,
Mikhail V. Puzanov.
I installed Postgresql from RedHat 5EL RPMS from the site. When I try to start
it as a service, it fails and pgstartup.log contains one record per startup
attempt stating: "runuser: warning: cannot change directory to /dev/null: Not a
directory."
Thanks!
---
So what is the state-of-the-art in the Postgresql world if I _do_ want
synchronous replication? 2-phase commit from the client application? Any
success/horror stories about doing it in Java?
---(end of broadcast)---
TIP 6: explain analyze is your
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED] On Behalf Of
> Richard Broersma Jr
> Sent: Monday, November 26, 2007 10:28 AM
> To: Joshua D. Drake
> Cc: pgsql-general@postgresql.org
> Subject: Re: [GENERAL] Primary Key
>
> --- On Mon, 11/26/07, Joshua D. Drake <
In the high-availabilty situation with a warm standby, is it possible (or
planned) to be able to make standby readable?
This is a new feature in Oracle 11 and it is very important for a project I am
working on.
---(end of broadcast)---
TIP 6: expla
Hello,
We are considering using Postgresql for rather large project and I have
questions about where it stands in respect to the following.
Consider these two features already found in major commercial products.
A) "resource governor" - the ability to configure how much resources (IO per
seco
be _very_ slow.
It is a classical multitask bottleneck problem.
Regards,
Mikhail Terekhov
---(end of broadcast)---
TIP 3: if posting/reading through Usenet, please send an appropriate
subscribe-nomail command to [EMAIL PROTECTED] so that your
Hi all!
May I use function with parameters with triggers?
When I try it I have error: "Can't find function ..".
Is it normal?
Postre can't find my function. Why?
create function mytest(text) returns opaque as '
declare
parma1 like as $1;
begin
return NEW;
end;
' languag
Matt Friedman wrote:
>
> I have it and have found it very useful.
I think charachter about Functions and Triggers too small.
Server-side functions are very impotant for check input information and
integrity of database.
Best regards,
Mik
Hi,
Is it possible to run VACUUM ANALYZE while other backends are running,
or should I shutdown them before vacuuming?
Regards,
Mikhail
RETURN NULL;
END;
' LANGUAGE 'plpgsql';
CREATE TRIGGER notify_trigger AFTER INSERT OR UPDATE ON test FOR EACH ROW
EXECUTE PROCEDURE send_notify();
But when this procedure is called I'm getting an error:
ERROR: parser: parse error at or near "$1"
What is wrong here?
Regards,
Mikhail
t;C"
LC_NUMERIC="C"
LC_TIME="C"
LC_COLLATE="C"
LC_MONETARY="C"
LC_MESSAGES="C"
LC_ALL=
Any help, tip or hint would be appreciated.
Thank you, Mikhail.
23 matches
Mail list logo
