T results.customerid, c.customername, count(distinct
count1) AS "TODAYS_USERS", count(distinct count2) AS "LAST 7
DAYS" , count(distinct count3) AS "LAST 30 DAYS"
FROM (SELECT distinct case when s.modified >=
'2010-02-11' then s.modelnumber else null e
When in doubt - test.
Why not remove index in MySQL (or create index in PostgreSQL) and see
what happens.
Why trying compare "apples and oranges"?
Igor Neyman
> -Original Message-
> From: Yang Zhang [mailto:yanghates...@gmail.com]
> Sent: Monday, February 22, 2010 1:
Joe,
What PG version are running?
8.2 here complains when running your example:
ERROR: column foo.name does not exist
LINE 6: select foo.name from foo;
^
** Error **
ERROR: column foo.name does not exist
SQL state: 42703
Igor Neyman
> -Original Mess
ut couldn't find
any reference to this fix beeing backported into 8.2.
So, was it or was it not backported into 8.2.5?
Magnus, sorry for sending this message to you directly (as well as to
the list), but it's a really "hot" issue for me.
Regards,
Igor Neyman
What's the complete error message?
Vacuum is using maintenance_work_mem. What is your setting
maintenance_work_mem compared to your RAM size.
Igor Neyman
> -Original Message-
> From: akp geek [mailto:akpg...@gmail.com]
> Sent: Wednesday, March 03, 2010 1:10 PM
>
ble name');
Is there any way to do this with stock PostgreSQL, or via something in
contrib?
Would pgsql's HTAB with hash_create and hash_search be a good basis for a
simple C extension which would provide such feature ?
I found this to be quite useful in e.g. pl/pgsql triggers
See attached attached OneClick_PG_Installer notes.
Igor Neyman
> -Original Message-
> From: John Gage [mailto:jsmg...@numericable.fr]
> Sent: Friday, April 02, 2010 3:44 AM
> To: sachin.srivast...@enterprisedb.com
> Cc: pgsql-general@postgresql.org
> Subject: Re: &
l us, what business entity/rule you are trying to
implement here, it'll be easier to help you.
Igor Neyman
> -Original Message-
> From: Dino Vliet [mailto:dino_vl...@yahoo.com]
> Sent: Saturday, April 03, 2010 7:32 AM
> To: r...@iol.ie
> Cc: pgsql-general@postgresql.org
I saw such behavior a few years ago on multiple very busy databases
connected to the same EMC SAN. The SAN's cache got overwhelmed by the
databases IO, and the storage latency went up significantly. I don't
remember now what was the latency, but it was above 40ms.
Is everything ok with your storag
I'm in a similar position, cloning a database multiple times to provide
development and qa databases for multiple groups. A feature desired by
Clark would greatly help many people with their development and qa
databases.
On the other hand, I imagine, a feature like this would not be easy to
develo
sell a little cheaper, and he who considers price only is that man's lawful
prey."
>
>
I might be a bit late in this discussion.
But I think, your problem is right here:
" running VACUUM FULL pg_largeobject"
If you are running "VACUUM FULL ..." on the tab
> -Original Message-
> From: Stefan Keller [mailto:sfkel...@gmail.com]
> Sent: Friday, January 06, 2012 1:12 AM
> To: Igor Neyman
> Cc: Simon Windsor; pgsql-general@postgresql.org
> Subject: Re: Vacuum and Large Objects
>
> Hi Igor
> 2011/12/16 Igor Neyman w
Stefan,
> -Original Message-
> From: Stefan Keller [mailto:sfkel...@gmail.com]
> Sent: Friday, January 06, 2012 1:12 AM
> To: Igor Neyman
> Cc: Simon Windsor; pgsql-general@postgresql.org
> Subject: Re: Vacuum and Large Objects
>
> Hi Igor
> 2011/12/16 Igor N
t seems that having synchronized
transaction id between let say streaming-replicated databases would give
some advantages if done properly.
Regards
Igor Polishchuk
--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql
Actually, what works is
set search_path='crabdata', 'public' ;
On 2/23/12 1:10 PM, "Adrian Klaver" wrote:
> On 02/23/2012 01:08 PM, Willem Buitendyk wrote:
>> I have it set in postgresql.conf and I've also used:
>> alter user postgres set search_path = crabdata,public;
>>
>
> Well search_pat
en dealing with that many connections.
For last 5 years we are running PG on many, many (hundreds) servers
under WIN, and we have no complains regarding PG stability.
Regards,
Igor Neyman
From: Serge Fonville [mailto:serge.fonvi...@gmail.com]
Sent: Friday, February 24, 2012 11:07 AM
To: mgo..
n PL/Proxy to remove the sequence from the argument list and to
call the sequence generator?
Is there a better way to do that?
Thanks in advance,
Igor Katson.
--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.or
Thom Brown wrote:
I would like to see what's in a function. I can do this if I use
pgAdmin III just by clicking on the function, and it appears in the
SQL pane opposite. But how is this done? Is there a command like
DESCRIBE FUNCTION logging.do_stuff; ? And if possible, doing this
without
I want to DROP CASCADE a table, but I am afraid that amoung numerous
recursive dependencies there will be smth, that I don't want to drop.
Is there a way to watch all dependencies recursively without doing a drop?
--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make cha
Thom Brown wrote:
I would like to see what's in a function. I can do this if I use
pgAdmin III just by clicking on the function, and it appears in the
SQL pane opposite. But how is this done? Is there a command like
DESCRIBE FUNCTION logging.do_stuff; ? And if possible, doing this
without
Is there a way to get i.e. table creation sql script from an existing
table in psql (not postgresql, but psql client), like it is in pgAdmin?
I.e. i point it to existing table 'foo', and it writes:
CREATE TABLE foo (
bar int
);
--
Sent via pgsql-general mailing list (pgsql-general@postgresql.
Grzegorz Jaśkiewicz wrote:
pg_dump -t foo database
Thanks, but pg_dump is not psql client (i meant the */bin/psql
interactive shell), and there is only an option for table objects, and
no one for i.e. indices.
--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make ch
Grzegorz Jaśkiewicz wrote:
pg_dump -t ANYOBJECT database
afaik., try it - play with it.
that does not work for indices. But the index creation is shown when
placing it's parent table into -t. Thanks for the help, Grzegorz, the
issue is solved.
--
Sent via pgsql-general mailing list (pgsq
Richard Huxton wrote:
Igor Katson wrote:
I want to DROP CASCADE a table, but I am afraid that amoung numerous
recursive dependencies there will be smth, that I don't want to drop.
Is there a way to watch all dependencies recursively without doing a drop?
BEGIN;
DROP CA
vidual, which are in a given group. That
seems ok without plproxy, but with using it, I can't imagine how can I
form a nice query, or a function (or a set of plpgsql + plproxy
functions) to do the job.
Thanks in advance and regards,
Igor Katson.
--
Sent via pgsql-general mailing list
Hannu Krosing wrote:
On Thu, 2009-01-22 at 02:33 +0300, Igor Katson wrote:
So to say, give me the list of friends (not only their ID's, but all the
needed columns!) of given individual, which are in a given group. That
seems ok without plproxy, but with using it, I can't imagine
Marko Kreen wrote:
On 1/22/09, Igor Katson wrote:
Hannu Krosing wrote:
> On Thu, 2009-01-22 at 02:33 +0300, Igor Katson wrote:
>
>
>> So to say, give me the list of friends (not only their ID's, but all the
>> needed columns!) of given individual, which
Hannu Krosing wrote:
On Thu, 2009-01-22 at 15:10 +0300, Igor Katson wrote:
Ok, thank you, guys. What is the best way to make an array out of a
column? I didn't make up anything better then writing a function:
CREATE OR REPLACE FUNCTION int_column_to_array(query text) RETURNS int[
Hannu Krosing wrote:
On Thu, 2009-01-22 at 16:13 +0200, Hannu Krosing wrote:
On Thu, 2009-01-22 at 15:10 +0300, Igor Katson wrote:
Ok, thank you, guys. What is the best way to make an array out of a
column? I didn't make up anything better then writing a function:
CREATE OR RE
statement will be removed, so if it was not written.
I think, this is a common problem.
Thanks in advance and regards,
Igor Katson.
--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general
Raymond O'Donnell wrote:
On 23/01/2009 11:16, Igor Katson wrote:
How do I write a function without complex logic, which will do:
a) If the input argument is NULL, then the corresponding select
statement will change from
column = arg
to
column IS NULL
You could build your stat
Sam Mason wrote:
On Fri, Jan 23, 2009 at 02:16:34PM +0300, Igor Katson wrote:
a) If the input argument is NULL, then the corresponding select
statement will change from
column = arg
to
column IS NULL
I think you want to use the IS [NOT] DISTINCT FROM operator. It
works like the
Michael Glaesemann wrote:
On Jan 23, 2009, at 10:11 , Igor Katson wrote:
That one is awesome, thanks, I completely forgot about CASE
statement. The search func now looks as follows, and works perfectly:
CREATE OR REPLACE FUNCTION isocial_user_func.search_users
(i_city_id int, i_edu_id int
FROM isocial_user_func.get_users_from_array(arr) a,
>>friend_func.get_friends_short(i_user_id,
limit_, offset_) b
WHERE a.id = b.friend_id
LOOP
RETURN NEXT rec;
END LOOP;
RETURN;
END;
$$ language plpgsql;
Regards,
Ig
postgres' full-text search capabilities?
2. Does anyone know, if there is a Russian dictionary of these names?
3. What is the name of such kind of dictionary, to know, what to google for?
Thanks in advance, and regards
Igor Katson.
--
Sent via pgsql-general mailing list (pgsql-general@postgresql.or
Oleg Bartunov wrote:
On Thu, 29 Jan 2009, Igor Katson wrote:
I have a column, containing the name of the user and there is a need
to organize an indexed search on this column. As far as I understand,
I need to use the full-text search capabilities of postgres.
I would like to attach a
As far as I understand, it is a known problem of using plproxy, that it
cannot be rolled back if used inside transactions. But I need something
similar to this functionality.
I have some data, that is duplicated across the DB partitions, and to be
exact, there is i.e. a plproxy-partitioned DB,
Hannu Krosing wrote:
On Fri, 2009-01-30 at 14:09 +0300, Igor Katson wrote:
As far as I understand, it is a known problem of using plproxy, that it
cannot be rolled back if used inside transactions. But I need something
similar to this functionality.
I have some data, that is duplicated
Oleg Bartunov wrote:
On Tue, 3 Feb 2009, Igor Katson wrote:
On my question about doing a first name search by it's synonyms, like
William Will Bill Billy
James Jim Jimmy
Oleg answered, that I can use dict_xsyn for that.
In the dict_xsyn manual, it is said that:
"This dicti
Oleg Bartunov wrote:
I contacted with author of dict_xsyn, so probably, he'll add option to
support what you want.
Thanks for cooperation, Oleg. By now, I will use that workaround.
--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http:/
On my question about doing a first name search by it's synonyms, like
William Will Bill Billy
James Jim Jimmy
Oleg answered, that I can use dict_xsyn for that.
In the dict_xsyn manual, it is said that:
"This dictionary type replaces words with groups of their synonyms, and
so makes it
possibl
I am doing an ALTER TABLE to create a foreign key, however with some
rows i get:
insert or update on table "name" violates foreign key constraint "name_fkey"
How can I just drop the conflicting rows while doing that?
--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make
Craig Ringer wrote:
Richard Huxton wrote:
DELETE FROM table1 WHERE col1 NOT IN (SELECT id from table2);
Just as a side note: If you have a large number of missing IDs and don't
want to wait a long time, you may be better off with something like
(untested, but I think it's right - TEST
I would like to call several plproxy functions one after another (which
will call plpgsql functions in different target partitions), and in case
one of them fails, i want to roll back changes in every one.
That is exactly how two-phase-commit (2PC) should work.
As far as I understand, the post
I need to execute a rather complex plpgsql function on a production
server, the result of which is hard to test by myself.
I want to make the result of a transaction, in which this function is
executed, visible to the outer clients, but to have the ability of
making a rollback in case one of t
Jeff Davis wrote:
On Wed, 2009-02-11 at 12:43 +0300, Igor Katson wrote:
ERROR: XX000: SPI_execute_plan failed executing query "PREPARE
TRANSACTION '2pctest'": SPI_ERROR_TRANSACTION
It's probably treating the word PREPARE specially. You can avoid this
pro
Jeff Davis wrote:
On Thu, 2009-02-12 at 00:57 +0300, Igor Katson wrote:
Thanks, Jeff. That's not good news, cause I am not able to do that.
There may be some creative solution, but I don't know plproxy well
enough to suggest one.
The postgres manual says, that
&quo
For each element in the array, I need to make some operation with plpgsql.
I usually use the syntax:
DECLARE
array_len int;
BEGIN
array_len := array_upper(i_array, 1);
FOR i IN 1 .. array_len
LOOP
SOME OPERATION (i_array[i])
END LOOP;
But I don't like that. Is there any built-in way to
A. Kretschmer wrote:
In response to Igor Katson :
I think I need a built-in function to make a column from an array, like
in the backwards operation SELECT ARRAY(column)
By David Fetter:
CREATE OR REPLACE FUNCTION unnest(ANYARRAY) RETURNS SETOF ANYELEMENT
LANGUAGE SQL AS $$SELECT $1
Is there any kind of a parallel psql client, to control a cluster of
databases, without the need to write a plproxy function for each request?
--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-gene
Sophie,
The sql like this:
select * from tbl1 where (a, b) in ((1, 20), (2, 30), (3, 50));
works very well in PostgreSQL 8,
Sincerely,
Igor Katrayev, Data Systems Manager
North Pacific Research Board
1007 West Third Avenue, Suite 100
Anchorage, AK 99501
Phone: 907-644-6700
Fax: 907-644-6780
was 8.1.2, and the bug appeared on v8.1.2; I've compiled and installed v8.1.5
over it. Autovacuum is enabled.
How can I fix this ?
--
Best Regards,
Igor Shevchenko
---(end of broadcast)---
TIP 4: Have you searched our list archi
ta" or run initdb
with an argument other than "/usr/pgsql/data".
--
Best Regards,
Igor Shevchenko
---(end of broadcast)---
TIP 5: don't forget to increase your free space map settings
Hi Merlin,
On Tuesday 14 November 2006 17:36, you wrote:
> On 11/13/06, Igor Shevchenko <[EMAIL PROTECTED]> wrote:
> > I've got a problem w/ one of pgsql installations. It can't start:
> >
> > [EMAIL PROTECTED] ~]$ /usr/pgsql/bin/postgres -D /usr/pgsql/data
On Tuesday 14 November 2006 17:34, Alvaro Herrera wrote:
> [EMAIL PROTECTED] wrote:
> > He's just trying to do what you told him.
> > But the use of initdb won't correct the problem Igor got. There's
> > something
> >
> > wrong with his databa
On Wednesday 15 November 2006 01:41, you wrote:
> On Tue, 14 Nov 2006 21:26:02 +0200, Igor Shevchenko wrote:
> >> I'd think starting a standalone backend and issuing a "reindex database"
> >> should be enough to get him started. Now, the problem is figuring _h
ctory "/usr/pgsql/data" or run initdb
> > with an argument other than "/usr/pgsql/data".
>
> of course if want to use /usr/pgsql/bin/initdb -D /usr/pgsql/data
> directory /usr/pgsql/data must be empty and user postgres must be the
> owner, but I don
read;
In the 1st case, partial index was used for the first query.
--
Best Regards,
Igor Shevchenko
---(end of broadcast)---
TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]
ction by regprocedure from
pl/pgsql. Is there any way to do it except writing my own "C" function ? Or
maybe I'd better off adding a "text" field for function name and calling it
from pl/pgsql with EXECUTE "string" ?
--
Best Regards,
Igor Shevchenko
--
l only but :
DROP SEQUENCE category_category_id_seq - fails saying that table category
column category_id uses it again
Any suggestions?
Thank you,
Igor
---(end of broadcast)---
TIP 5: Have you checked our extensive FAQ?
another (manually). Is there some quick tool
for doing this ?
There was a thread about it sometime in aug, 2002, but it ended without
producing anything useful.
--
Best regards,
Igor Shevchenko
---(end of broadcast)---
TIP 5: Have you ch
rcass.ath.cx/diffpg.pl
It doesn't honor schema and object ownership.
--
Best regards,
Igor Shevchenko
---(end of broadcast)---
TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]
e's attributes) into TABLE [(list of
attributes)] values (list of values)
I dont know if there's anything about this topic in the SQL spec; those are
just my random thoughs. Personally, I've had several situations where I'd
benefit from having this feature in either
he database was via insert-or-update path.
--
Best regards,
Igor Shevchenko
---(end of broadcast)---
TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]
it a relational or a non-relational
task ?
Are there any plans on adding this feature per sql 2003 spec ?
--
Best regards,
Igor Shevchenko
---(end of broadcast)---
TIP 6: Have you searched our list archives?
http://archives.postgresql.org
EXT: PL/pgSQL function "add_messages_to_folder" line 9 at fetch
I've tried to send this query with and without bound parameter, with no luck
(same error message).
Can someone shed some light on this ? :)
--
Best regards,
Igor Shevchenko
---(end of broa
is that yours?
--- Trend GateLock [EMAIL PROTECTED] (主機：higp3.gatelock.com.tw)
** 中毒檔案 shower.com 已刪除。
Trend GateLock [EMAIL PROTECTED] (主機：higp3.gatelock.com.tw)
** 在檔案 shower.com 中發現病毒 WORM_NETSKY.C。
無法清除病毒，中毒檔案已刪除。
.) returns 1015 for those columns. I couldn't find any #define for
this OID -- does this mean it can change in future pg releases ? Also, can I
safely use the binary format as described (by the code) in "array_send" ? Is
this interface is frozen, as a part of protocol v3
Thank you very much.
This works:
UPDATE
test
SET
code=T2.code
FROM
test T2
WHERE
test.code=0 AND
test.master=T2.name;
Igor
---(end of broadcast)---
TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]
U postgres template1 < db_backup.sql was
executed.
I want to underline not all "timestamp default now()" columns but only some
of them.
Any ideas?
Thank you,
Igor
---(end of broadcast)---
TIP 3: if posting/reading through U
TIME ZONE 'EUROPE/MOSCOW'
Tell me please , what I made wrong?
Thanks for any suggestions.
Igor.
On Fri, Oct 27, 2000 at 03:01:31PM -0400, Tom Lane wrote:
> This gives you back an integer number of seconds per Unix conventions,
> which you just use atoi() on. For example:
>
> regression=# select date_part('epoch', now());
Thanx
--
Igor Roboul, Unix System Admi
On Fri, Oct 27, 2000 at 07:12:57PM -0200, Vilson farias wrote:
>I start postgres with these params: su -l postgres -c
> '/usr/bin/postmaster -B 2048 -i -D "/home/postgres/data"' &.
Try starting postmaster with 'nice':
nice /usr/bin/postmaster -B 2048
ink(cnt,fid,wid) values(1,v_fid,v_wid);"
inserts new record, or just adds 1 to counter.
--
Igor Roboul, Unix System Administrator & Programmer @ sanatorium "Raduga",
Sochi, Russia
http://www.brainbench.com/transcript.jsp?pid=304744
quot;
Try removing first "GROUP BY messages.msgid HAVING count(messages.msgid)=2)"
--
Igor Roboul, Unix System Administrator & Programmer @ sanatorium "Raduga",
Sochi, Russia
http://www.brainbench.com/transcript.jsp?pid=304744
On Thu, Nov 02, 2000 at 05:35:04PM +0600, Denis Perchine wrote:
> Except on case... When you would like to be sure in transaction safety...
Ok, but not for image galery.
--
Igor Roboul, Unix System Administrator & Programmer @ sanatorium "Raduga",
Sochi, Russia
http://w
On Thu, Nov 02, 2000 at 02:39:54PM +0300, Igor Roboul wrote:
> On Thu, Nov 02, 2000 at 05:35:04PM +0600, Denis Perchine wrote:
> > Except on case... When you would like to be sure in transaction safety...
> Ok, but not for image galery.
I have answered on argument about transaction
ern.ipc.shmmni: 96
kern.ipc.shmseg: 64
kern.ipc.shmall: 1024
kern.ipc.shm_use_phys: 0
--
Igor Roboul, Unix System Administrator & Programmer @ sanatorium "Raduga",
Sochi, Russia
http://www.brainbench.com/transcript.jsp?pid=304744
y_key_field'. I think.
--
Igor Roboul, Unix System Administrator & Programmer @ sanatorium "Raduga",
Sochi, Russia
http://www.brainbench.com/transcript.jsp?pid=304744
On Mon, Nov 20, 2000 at 12:44:20PM +0800, BaiJie wrote:
> I KNOW THERE IS A OID FIELD ,BUT IT 'S WIRED , NOT A INTEGER INDEX FROM I TO
> CURRENT ROW NUMBER!! HOW CAN I GET THE TRUE ROW NUMBER IN POSTGRESQL!!
At first, don't use all caps.
--
Igor' Robul', Unix System
On Fri, Nov 24, 2000 at 08:35:34AM +0300, Igor' Robul' wrote:
> On Mon, Nov 20, 2000 at 12:44:20PM +0800, BaiJie wrote:
> > I KNOW THERE IS A OID FIELD ,BUT IT 'S WIRED , NOT A INTEGER INDEX FROM I TO
> > CURRENT ROW NUMBER!! HOW CAN I GET THE TRUE ROW NUMBER IN POST
text
is not very good, you can store big text into BLOB
> might be extended for storing binary data as well (e.g. large images).
For binary data use BLOBs
--
Igor' Robul', Unix System Administrator & Programmer @ sanatorium "Raduga",
Sochi, Russia
http://www.brainbench.com/transcript.jsp?pid=304744
ome, etc.), since I imagine it depends on the fact that
the port is open.
My question is therefore:
1. is there a way to permit the pl/python to connect to the internet all
the time and with a better configuration?
2. or should I forget the procedure and write a python script outside
the database?
nnect to the internet. It works very unreliable.
Have you maybe other hints?
Many thanks in advance.
--
:: Igor Sosa Mayor :: joseleopoldo1...@gmail.com ::
:: GnuPG: 0x1C1E2890 :: http://www.gnupg.org/ ::
:: jabberid: rogorido ::::
--
Sent via pgsql
g/10366558
Then a trigger uses this code to insert the coordinates in a table.
Maybe I'm doing something wrong...
--
:: Igor Sosa Mayor :: joseleopoldo1...@gmail.com ::
:: GnuPG: 0x1C1E2890 :: http://www.gnupg.org/ ::
:: jabberid: rogorido ::::
the impression postgres is trying to make a connection
without having to rights to get the answer (I'm on linux). I dont
understand exactly which process (and with which rights) is executing
the plpython procedure...
Thanks in any case for your answer.
--
:: Igor Sosa Mayor :: joseleopo
John R Pierce writes:
> On 8/22/2015 12:23 AM, Igor Sosa Mayor wrote:
>> Somehow I have the impression postgres is trying to make a connection
>> without having to rights to get the answer (I'm on linux). I dont
>> understand exactly which process (and with which r
ection...
http://pastie.org/10368578
--
:: Igor Sosa Mayor :: joseleopoldo1...@gmail.com ::
:: GnuPG: 0x1C1E2890 :: http://www.gnupg.org/ ::
:: jabberid: rogorido ::::
--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes
Adrian Klaver writes:
> Are you taking lack of coordinates in the table as evidence of no
> connection or are you determining that some other way?
thanks again for your answer. The evidence is very easy. I get this[1].
BUT: if I try exactly the same python code in a console there is no
error and
"Charles Clavadetscher" writes:
> Are you sure that you really use exactly the same code? The indication
> "survice unknown" seems to point to an error in the URL and not to an
> authorization problem. Maybe some encoding problem?
THnaks for your hints.
Yes, exactly the same code. Exactly the s
net when
I use the function. I dont see any packages. From the python console
there is a connection to nominatim (openstreetmap).
I think there is exactly the problem: PG does not make any attempt (or
it does but it fails even before any package is being sent) to the
internet.
THnaks for your ideas
Adrian Klaver writes:
> What happens if you run the function below directly in psql and not
> through the trigger?
exactly the same error.
> Do you have both Python 2 and Python 3 on this machine?
I have both.
> Did you create both the 2u and 3u variants of plpythonu?
I tried both... in my
reach the remote service.
i don't understand exactly (it is the 1. time I write a procedure even
in PG). It is so:
1. I run ipython as my user (igor);
2. I run postgres both as user and as postgres. In both cases the same
error. PG does not run ipython, because ipython is just a console fo
Adrian Klaver writes:
> I tried it here, using your functions, and could not get it to fail:
well... i'm happy my function is working... my 1. procedure in PG... Now
maybe I'm able to get it working in my laptop...
> So it is something to do with the setup on your laptop.
yes, or the network I
Adrian Klaver writes:
>> I tried both... in my system (archlinux) there is plpython2u and
>> plpythonu. I think the 2nd one is python3.
>
> How did they get there?
>
> Or more to the point how where they built?
just installed with postgresql which comes with archlinux. I did not
built anything.
Adrian Klaver writes:
> What is the default system Python?
python3. But see my other answer. I dont think this is the problem.
--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general
Igor Sosa Mayor writes:
> My question is therefore:
> 1. is there a way to permit the pl/python to connect to the internet all
>the time and with a better configuration?
> 2. or should I forget the procedure and write a python script outside
>the database?
I want to updat
Igor Sosa Mayor writes:
> Igor Sosa Mayor writes:
>
>> My question is therefore:
>> 1. is there a way to permit the pl/python to connect to the internet all
>>the time and with a better configuration?
>> 2. or should I forget the procedure and write a p
Adrian Klaver writes:
>> But I see that the log options in PG are really rich. Could maybe
>> someone tell me which could be the best options to find the problem? I
>> will be offline now during 24h, but I will try to make some experiments
>> in the meantime.
>
> From here:
> http://www.postgresq
Igor Sosa Mayor writes:
> Maybe it has something to do with systemd? I'm trying to get all debug
A little more information: the unit of postgresql in my systemd looks
like this[1]. That means, it is started of course after the network (but
maybe there is not any connection avalaib
401 - 500 of 505 matches
Mail list logo
