On 08/07/2012 11:51 AM, Shridhar Daithankar wrote:
testdb2=# revoke connect ON database testdb2 FROM testuser1;
REVOKE
You can't revoke a permission that isn't set. PostgreSQL doesn't have
explicit deny rules, so you can only remove a grant.
The documentation on databases doesn't seem to
(Reposted as the list manager appears to have eaten the first copy):
Hey all
It seems to be surprisingly hard to build JSON structures with
PostgreSQL 9.2's json features, because:
- There's no aggregate, function or operator that merges two or more
objects; and
- there's no single-value
Shridhar Daithankar wrote:
I am trying to setup a cluster for trac databases and want to isolate
each db, by assigning a specific
user to a DB.
I followed the documentation but as shown in the following example,
limiting access by connect does
not seem to be working.
What am I missing?
Hi Samba,
first: do not worry, it is perfectly normal.
On Tue, 7 Aug 2012 16:25:14 +0530, Samba saas...@gmail.com wrote:
Hi all,
I'm seeing some weired errors in the postgres logs after upgrading to
postgres-9.1(.3) about the schema added by default to search patch
WARNING: invalid value
Hi,
I've implemented an aggregation function to compute quartiles in C
borrowing liberally from orafce code. I uses this code in a windowing
context and it worked fine until today - and I'm not sure what
changed. This is on 9.1.2 and I have also tried it on 9.1.4.
What I have determined so far
Hello.
I've just discovered a very strange thing:
SELECT '1 mon'::interval = '30 days'::interval -- TRUE???
This returns TRUE (also affected when I create an unique index using an
interval column). Why?
I know that Postgres stores monthes, days and seconds in interval values
separately. So
Dnia 6 sierpnia 2012 17:00 Adrian Klaver adrian.kla...@gmail.com napisał(a):
The clog has somewhat re-formed - the full listing of lsof (filtered for
unique files) for postmaster(s) on the database mount is here:
http://BillionUploads.com/ya9kjv78t9es/postmaster_files_sorted.csv.html
...and even worse:
SELECT ('1 year'::interval) = ('360 days'::interval); -- TRUE :-)
SELECT ('1 year'::interval) = ('365 days'::interval); -- FALSE :-)
On Tue, Aug 7, 2012 at 4:42 PM, Dmitry Koterov dmi...@koterov.ru wrote:
Hello.
I've just discovered a very strange thing:
SELECT '1
On Mon, Aug 6, 2012 at 6:54 PM, Rodrigo Gonzalez lis...@estrads.com.ar wrote:
On 06/08/12 13:31, Bruce Momjian wrote:
For longer terms, perhaps we should set up an URL forwarder or
something that the docs can link through in the cases where we really
need this, so we can more easily update the
Hey Everyone,
I've got a bit of an interesting issue going on with pg_xlog growing on
a streaming replication slave.
We're running postgres 9.1.1 x64 built from source on Centos 5.8 x64.
On both the master and the slave we have wal_keep_segments configured for
1000
wal_keep_segments = 1000
Thanks Gabriele for those pointers,
I could now narrow it down to two things:
1. system_data user logging into other databases [one of those may be
the default 'postgres'] which does not have system_data schema
2. other users [like 'postgres'] logging into their own or even other
We run several instances of postgre in different countries, and we try keeping
them as same as possible, in terms of structure of the tables and function
definitions (except the content of schema config, which differs between dbs).
So if we need to implement some different algorithm per
Samba saas...@gmail.com writes:
I'm seeing some weired errors in the postgres logs after upgrading to
postgres-9.1(.3) about the schema added by default to search patch
WARNING: invalid value for parameter search_path: system_data
DETAIL: schema system_data does not exist
...
Could anyone
Hello
Now my questions is: Are the stored functions (both plpgsql and plain sql
functions) kept always in a memory? Or they are stored similarly like
tables, on the disk, reading them into memory when called and possibly
release them from memory, if memory is needed for something else?
Samba wrote:
I'm seeing some weired errors in the postgres logs after upgrading
to postgres-9.1(.3) about the schema added by default to search patch
WARNING: invalid value for parameter search_path: system_data
DETAIL: schema system_data does not exist
We do have a user named
Hi all,
I now realize that the issue is indeed occurring when users who have
system_data in their search_path log in to other databases that does not
have that schema.
Could someone explain how to add schema(s) into search_path for a database
[not to user/role] irrespective of whichever user
On Tue, Aug 7, 2012 at 7:15 AM, Samba saas...@gmail.com wrote:
Thanks Gabriele for those pointers,
I could now narrow it down to two things:
system_data user logging into other databases [one of those may be the
default 'postgres'] which does not have system_data schema
other users [like
Adriaan Joubert adriaan.joub...@gmail.com writes:
I've implemented an aggregation function to compute quartiles in C
borrowing liberally from orafce code. I uses this code in a windowing
context and it worked fine until today - and I'm not sure what
changed. This is on 9.1.2 and I have also
Hi,
Finally got this running under the debugger and figured out what is
going on. I had been under the impression that
if (PG_ARGISNULL(0))
PG_RETURN_NULL();
state = (quartile_state *) PG_GETARG_POINTER(0);
would ensure that state was never a null pointer.
I have a field which contains an interval value and I sometimes need to
represent the full interval (not a part) as a decimal number of months. For
example, 5 years 6 mons 3 days as 66.1 months. I've been trying to figure
out how to do this and haven't found a definitive answer.
The
On Aug 7, 2012, at 8:41 AM, Aram Fingal fin...@multifactorial.com wrote:
I have a field which contains an interval value and I sometimes need to
represent the full interval (not a part) as a decimal number of months. For
example, 5 years 6 mons 3 days as 66.1 months. I've been trying to
Surprised to see this isn't offered as a Foreign Data Wrapper- one to other
Postgres servers. I was attempting to replace some uses I have of dbilink, and
found a couple places where I am using it to connect to Postgres. One is for
pseudo Autonomous Transactions- a db link to the same postgres
On Tue, Aug 7, 2012 at 3:20 AM, Craig Ringer ring...@ringerc.id.au wrote:
(Reposted as the list manager appears to have eaten the first copy):
Hey all
It seems to be surprisingly hard to build JSON structures with PostgreSQL
9.2's json features, because:
- There's no aggregate, function or
On Mon, Aug 6, 2012 at 3:29 AM, Ben Chobot be...@silentmedia.com wrote:
On Aug 5, 2012, at 11:12 AM, Fujii Masao wrote:
On Sat, Jul 28, 2012 at 2:00 AM, Ben Chobot be...@silentmedia.com wrote:
We make heavy use of streaming replication on PG 9.1 and it's been great for
us. We do have one
David Greco david_gr...@harte-hanks.com wrote:
Surprised to see this isn't offered as a Foreign Data Wrapper- one
to other Postgres servers.
People have been working on it. It seems quite likely to be
included in the 9.3 release next year.
-Kevin
--
Sent via pgsql-general mailing list
Kevin Grittner kevin.gritt...@wicourts.gov wrote:
David Greco david_gr...@harte-hanks.com wrote:
Surprised to see this isn't offered as a Foreign Data Wrapper- one
to other Postgres servers.
People have been working on it. It seems quite likely to be
included in the 9.3 release next
Great thanks. I see there is talk of 9.3 including autonomous transaction
support as well.
-Original Message-
From: Kevin Grittner [mailto:kevin.gritt...@wicourts.gov]
Sent: Tuesday, August 07, 2012 1:04 PM
To: David Greco; pgsql-general@postgresql.org
Subject: Re: [GENERAL] Feature
On Aug 7, 2012, at 9:32 AM, Fujii Masao wrote:
On Mon, Aug 6, 2012 at 3:29 AM, Ben Chobot be...@silentmedia.com wrote:
Oh, I would have though that doing a clean shutdown of the old master (step
1) would have made sure that all the unstreamed wal records would be flushed
to any connected
On Sun, Aug 5, 2012 at 10:12 PM, Fujii Masao masao.fu...@gmail.com wrote:
Have we just avoided running pg_basebackup, or have we just given ourselves
data corruption?
If you change your operations in the above-mentioned way, I think you can
avoid pg_basebackup on the planned switch. I've not
On Tue, Aug 7, 2012 at 11:31 AM, Merlin Moncure mmonc...@gmail.com wrote:
On Tue, Aug 7, 2012 at 3:20 AM, Craig Ringer ring...@ringerc.id.au wrote:
(Reposted as the list manager appears to have eaten the first copy):
Hey all
It seems to be surprisingly hard to build JSON structures with
Hi
select
case when somevariable = 2
then (insert into pipe (line)
select bob.edge_data.edge_id
from bob.edge_data, bob.node, pipe
where st_intersects(st_startpoint(bob.edge_data.geom), bob.node.geom)
and bob.node.node_id = 415
and pipe.id = 1)
I am
On Wed, Aug 8, 2012 at 8:26 AM, Bob Pawley rjpaw...@shaw.ca wrote:
Hi
select
case when somevariable = 2
then (insert into pipe (line) ...
I am attempting to use the above. However, with or without the enclosing
brackets I get a syntax error on the word into.
Utterly untested, but does
On 8 Aug 2012, at 24:26, Bob Pawley wrote:
Hi
select
case when somevariable = 2
then (insert into pipe (line)
select bob.edge_data.edge_id
from bob.edge_data, bob.node, pipe
where st_intersects(st_startpoint(bob.edge_data.geom), bob.node.geom)
and
Hi Alban
Probably no difference except I have four cases and I was trying, in an
attempt to save processing time, to compact commands a little.
Bob
-Original Message-
From: Alban Hertroys
Sent: Tuesday, August 07, 2012 3:40 PM
To: Bob Pawley
Cc: Postgresql
Subject: Re: [GENERAL]
From: pgsql-general-ow...@postgresql.org
[mailto:pgsql-general-ow...@postgresql.org] On Behalf Of Bob Pawley
Sent: Tuesday, August 07, 2012 6:26 PM
To: Postgresql
Subject: [GENERAL] Using Insert with case
Hi
select
case when somevariable = 2
then (insert into pipe (line)
select
Hi all,
I am Haiming, a software engineer.
One of our product is depending on PostgreSQL. We know the product code for 8.2
and 8.3 is {B823632F-3B72-4514-8861-B961CE263224}. Anyone who knows the
product code for postresql 9.1 could you please provide it and how can we find
the product code
On Wed, Aug 8, 2012 at 3:03 AM, Brian McNally bmcna...@uw.edu wrote:
[root@gvsdb-dev tmp]# gdb /usr/pgsql-9.0/bin/postmaster 1160
GNU gdb (GDB) Red Hat Enterprise Linux (7.0.1-32.el5_6.2)
(gdb) bt
#0 0x00378f8d5497 in semop () from /lib64/libc.so.6
#1 0x005bc1c3 in
On 08/08/2012 03:45 AM, Merlin Moncure wrote:
Given that you can do that, if you had the ability to emit json from
an hstore the OP's problem would be trivially handled.
That's where my thinking went at first too, but there's a wrinkle with
that: json represents the number 1 and the string 1
On 08/08/2012 08:17 AM, Haiming Zhang wrote:
Hi all,
I am Haiming, a software engineer.
One of our product is depending on PostgreSQL. We know the product code
for 8.2 and 8.3 is {B823632F-3B72-4514-8861-B961CE263224}. Anyone who
knows the product code for postresql 9.1 could you please
On Tuesday 07 Aug 2012 12:21:04 AM Tom Lane wrote:
Shridhar Daithankar ghodech...@ghodechhap.net writes:
I am wondering, why following two values result in a shift by 3.5 hours. I
would expect them to be identical.
I understand that canonical time zone names could be ambiguous at times
40 matches
Mail list logo