Re: [GENERAL] Error: absolute path not allowed

[Raghavendra]

On Thu, Jan 10, 2013 at 1:20 AM, wschwurack  wrote:

> How do I find what is calling pg_read_file
>
>
As said, check the log's to find DB,Host,User etc. from where its been
called. If log_line_prefix not set to appropriate to all those details then
change it as shown below which just need RELOAD of server. After changing
your logs should show those details. Eg:-
log_line_prefix = '%t [%p]: [%l-1] db=%d,user=%u client=%h'

Also, its just my guess, if pg_read_file() function called from any other
function then you can check with below query.

select proname from pg_proc where prosrc ilike '%pg_read_file%' and
pronamespace=(select oid from pg_namespace where nspname='public');

---
Regards,
Raghavendra
EnterpriseDB Corporation
Blog: http://raghavt.blogspot.com/

Re: [GENERAL] Database connections seemingly hanging

[Tom Lane]

fredrik.huitfeldtmad...@schneider-electric.com writes:
> We have a setup where 2 JBoss (5.1) servers communicate with 1 instance of 
> PgPool (3.04), which again communicates with 2 Postgresql (8.4) servers. 
> The JBoss servers host some Java code for us and as part of that they run 
> some quartz jobs. 

> These jobs are triggered right after startup and as part of that we get 
> what seems to get stuck. At least when we can see in the database that 
> when inspecting pg_locks, there exists a virtual transaction that has all 
> desired locks granted but seems to be stuck. When we inspect 
> pg_stat_activity, it seems that the process is still waiting for the query 
> (SELECT ... FOR UPDATE) to finish.

> The locking transaction is described here: http://pastebin.com/3pEn6vPe

What that shows is several sessions running SELECT FOR UPDATE, but none
of them seem to be waiting.  What else is going on?  In particular, are
there any idle-in-transaction sessions?  Also, would any of these
SELECTs return enough rows that the sessions might be blocked trying to
send data back to their clients?  That wouldn't show as waiting = true,
though I think you could detect it by strace'ing the backends to see if
they are stopped in a send() kernel call.

> We suspect that a connection to the database acquires its locks but 
> somehow does not return to the application. If this is true, it would 
> either be a postgresql or a pgpool problem. We would appreciate any help 
> in further debugging or resolving the situation. 

It seems like a good guess would be that you have a deadlock situation
that cannot be detected by the database because part of the blockage is
on the client side --- that is, client thread A is waiting on its
database query, that query is waiting on some lock held by client thread
B's database session, and thread B is somehow waiting for A on the
client side.  It's not too hard to get into this type of situation when
B is sitting on an open idle-in-transaction session: B isn't waiting for
the database to do anything, but is doing something itself, and so it's
not obvious that there's any risk.  Thus my question about what idle
sessions there might be.  This does usually lead to a visibly waiting
database session for client A, though, so it's probably too simple as an
explanation for your issue.  We have seen some harder-to-debug cases
where the database sessions weren't visibly "waiting" because they were
blocked on client I/O, so maybe you've got something like that.

Another line of thought to pursue is possible misuse of pgpool.  If
pgpool doesn't realize you're inside a transaction and swaps the
connection to some other client thread, all kinds of confusion ensues.

Also, I hope you're running a reasonably recent 8.4.x minor release.
A quick look through the commit logs didn't show anything about deadlock
fixes in the 8.4 branch, but I might have missed something that was
fixed a long time ago.

regards, tom lane


-- 
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general

Re: [GENERAL] Bug: dblink_send_query not work on 9.2?

[Tom Lane]

aasat  writes:
> This work

> do $$
> begin
>   perform dblink.dblink_connect('internal'); 

I believe that creates an unnamed connection referencing a foreign
server definition named "internal".

>   perform dblink.dblink_exec('internal', 'set application_name=''dblink'';',
> true); 

This is going to find that "internal" is not the name of any known
dblink connection (since the one that does exist is unnamed).  So it
creates a transient connection of its own using "internal" as a
connection string, ie a fresh reference to the foreign server.  For the
duration of this command, there are two connections to the remote
database.

>   perform dblink.dblink_disconnect();

And that closes the unnamed connection.

>   perform dblink.dblink_send_query('internal', 'select dummy();');

This is going to fail because there is no dblink connection named
"internal".  Unlike dblink_exec, there is no support for making a
transient connection for just the duration of the function call,
since it wouldn't be sensible --- the connection has to persist
so you can get the result back.

Use the two-parameter form of dblink_connect() so you can make a
named connection.

> It's works without problems on version 9.0

Consulting the commit logs, I see that that was a bug we fixed
as of 9.1.

Author: Joe Conway 
Branch: master Release: REL9_2_BR [8af3596d6] 2011-06-25 15:58:07 -0700
Branch: REL9_1_STABLE Release: REL9_1_0 [e4fb58f89] 2011-06-25 15:40:49 -0700

Async dblink functions require a named connection, and therefore should
use DBLINK_GET_NAMED_CONN rather than DBLINK_GET_CONN.
Problem found by Peter Eisentraut and patch by Fujii Masao.

regards, tom lane


-- 
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general

[GENERAL] Custom JDBC wrapper for DB2 Fed Server

[Alexander Gataric]

I was wondering if there is a Postgres JDBC Type 4 wrapper for DB2 Federated
Server. The default wrapper from IBM has a number of issues with Postgres
which a better wrapper could solve. I will attempt to modify the IBM wrapper
if I can get my hands on the source code.



-- 
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general

Re: [GENERAL] How can I detect if a schema exists?

[felix]

On Thu, Jan 10, 2013 at 09:11:59AM +0900, Ian Lawrence Barwick wrote:

> How about:
> 
>   SELECT TRUE FROM information_schema.schemata WHERE schema_name = 'xyzzy'
> 
> ? (Though I notice this table only seems to show the schemas owned by the
> user if the user is not a superuser).

I'll have to play with the users and permissions, but that sounds like
it could be a feature; the code wants to know if the user can use the
schema.

-- 
... _._. ._ ._. . _._. ._. ___ .__ ._. . .__. ._ .. ._.
 Felix Finch: scarecrow repairman & rocket surgeon / fe...@crowfix.com
  GPG = E987 4493 C860 246C 3B1E  6477 7838 76E9 182E 8151 ITAR license #4933
I've found a solution to Fermat's Last Theorem but I see I've run out of room o


-- 
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general

Re: [GENERAL] currval of sequence xxx_seq is not yet defined in this session?

[Adrian Klaver]

On 01/08/2013 05:58 PM, kenyon wrote:

Thanks for your reply!

yeath,I quite accept your opinion,once i guess the drive adds savepoint
between the two SQL，but not sure


Off hand I would say the drive has nothing to do with it. The issue 
would seem to be that as originally written ibatis was issuing two 
queries in different sessions. This meant the query trying to fetch the 
currval could not find it because the sequence nextval() was happening 
in another session.



--
Adrian Klaver
adrian.kla...@gmail.com


--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general

Re: [GENERAL] Error: absolute path not allowed

[Scott Marlowe]

Look at your log line prefix setting in postgresql.conf.  Specifically
log things like db, host, username, etc.

On Wed, Jan 9, 2013 at 12:50 PM, wschwurack  wrote:
> How do I find what is calling pg_read_file
>
>
>
> --
> View this message in context: 
> http://postgresql.1045698.n5.nabble.com/Error-absolute-path-not-allowed-tp5739320p5739424.html
> Sent from the PostgreSQL - general mailing list archive at Nabble.com.
>
>
> --
> Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
> To make changes to your subscription:
> http://www.postgresql.org/mailpref/pgsql-general



-- 
To understand recursion, one must first understand recursion.


-- 
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general

Re: [GENERAL] How can I detect if a schema exists?

[Ian Lawrence Barwick]

2013/1/10 

> I was reviving an old test program which hasn't been used since 9.1 and
> found that "SET search_path = xyzzy" no longer fails if the schema "xyzzy"
> doesn't exist.
>
> Is there an approved or recommended way to tell if a schema exists?  I can
> think of lots of ways, but none as easy as the 9.1 search_path.  I suppose
> I shouldn't rely on undocumented features ...
>

How about:

  SELECT TRUE FROM information_schema.schemata WHERE schema_name = 'xyzzy'

? (Though I notice this table only seems to show the schemas owned by the
user if the user is not a superuser).

Ian Lawrence Barwick

[GENERAL] Database connections seemingly hanging

[Fredrik . HuitfeldtMadsen]

Hi All,

We have a setup where 2 JBoss (5.1) servers communicate with 1 instance of 
PgPool (3.04), which again communicates with 2 Postgresql (8.4) servers. 
The JBoss servers host some Java code for us and as part of that they run 
some quartz jobs. 

These jobs are triggered right after startup and as part of that we get 
what seems to get stuck. At least when we can see in the database that 
when inspecting pg_locks, there exists a virtual transaction that has all 
desired locks granted but seems to be stuck. When we inspect 
pg_stat_activity, it seems that the process is still waiting for the query 
(SELECT ... FOR UPDATE) to finish.

The locking transaction is described here: http://pastebin.com/3pEn6vPe

We know that the quartz thread is attempting to obtain a row share lock. 
We know that we have enough connections available in postgres and in 
pgpool. We also know that the issue occurs much more frequently when we 
enable postgres statememt logging. We assume that this is due to postgres 
becomming slower as a result of the additionsl logging. When we look at 
the server thread dump, we can see that all quartz threads are either 
sleeping or waiting for postgres.

A thread dump of the relevant quartz threads is described here: 
http://pastebin.com/iPhuFLrM

It is important to note that the issue does not only occur with quartz 
jobs, but it is where we see it most frequently. This is likely to be due 
to the fact that it is the place where we have the highest level of 
concurrency.

We suspect that a connection to the database acquires its locks but 
somehow does not return to the application. If this is true, it would 
either be a postgresql or a pgpool problem. We would appreciate any help 
in further debugging or resolving the situation. 

Kind regards,
Fredrik

[GENERAL] Bug: dblink_send_query not work on 9.2?

[aasat]

I have problem with dblink_send_query function on Postgres 9.2.2

This work

do $$
begin
  perform dblink.dblink_connect('internal'); 
  perform dblink.dblink_exec('internal', 'set application_name=''dblink'';',
true); 
  perform dblink.dblink_disconnect();
end;
$$

But this not

do $$
begin
  perform dblink.dblink_connect('internal'); 
  perform dblink.dblink_send_query('internal', 'select dummy();');
  perform dblink.dblink_disconnect();
end;
$$

I got error

ERROR: connection "internal" not available
SQL state: 08003
Context: SQL statement "SELECT dblink.dblink_send_query('internal', 'select
dummy();')"
PL/pgSQL function inline_code_block line 5 at PERFORM

It's works without problems on version 9.0




--
View this message in context: 
http://postgresql.1045698.n5.nabble.com/Bug-dblink-send-query-not-work-on-9-2-tp5739365.html
Sent from the PostgreSQL - general mailing list archive at Nabble.com.


-- 
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general

Re: [GENERAL] Error: absolute path not allowed

[wschwurack]

How do I find what is calling pg_read_file



--
View this message in context: 
http://postgresql.1045698.n5.nabble.com/Error-absolute-path-not-allowed-tp5739320p5739424.html
Sent from the PostgreSQL - general mailing list archive at Nabble.com.


-- 
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general

Re: [GENERAL] How to store clickmap points?

[David Johnston]

aasat wrote
> Hi, 
> 
> I want to store clickmap points (X, Y and hits value) for website
> 
> I currently have table like this
> 
> CREATE TABLE clickmap (
>   page_id integer,
>   date date,
>   x smallint,
>   y smallint,
>   hits integer
> )
> 
> But this generated about 1M rows per day.
> 
> Can Postgres have better method to store this data? I also have the
> possibility to update hits value for point


Generally questions like this require that some knowledge of how the
resultant data is going to be used in order to make appropriate design
trade-offs.  I would also question why you wouldn't just store the actual
timestamp value as opposed to just storing the date.

David J.



--
View this message in context: 
http://postgresql.1045698.n5.nabble.com/How-to-store-clickmap-points-tp5739121p5739449.html
Sent from the PostgreSQL - general mailing list archive at Nabble.com.


-- 
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general

[GENERAL] How can I detect if a schema exists?

[felix]

I was reviving an old test program which hasn't been used since 9.1 and found 
that "SET search_path = xyzzy" no longer fails if the schema "xyzzy" doesn't 
exist.

Is there an approved or recommended way to tell if a schema exists?  I can 
think of lots of ways, but none as easy as the 9.1 search_path.  I suppose I 
shouldn't rely on undocumented features ...

-- 
... _._. ._ ._. . _._. ._. ___ .__ ._. . .__. ._ .. ._.
 Felix Finch: scarecrow repairman & rocket surgeon / fe...@crowfix.com
  GPG = E987 4493 C860 246C 3B1E  6477 7838 76E9 182E 8151 ITAR license #4933
I've found a solution to Fermat's Last Theorem but I see I've run out of room o


-- 
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general

Re: [GENERAL] [Solved] Corrupt indexes on slave when using pg_bulkload on master

[Tom Lane]

James Cowell  writes:
> But pg_bulkload only puts the index updates into WAL if you also have 
>  
> archive_mode = on
>  
> I guess it needs to test wal_level rather than archive mode now?  It looks 
> like changes to the project have been minimal for some time, which is a shame 
> because it's a very useful tool.

Oh, the code in question is in pg_bulkload not the backend?  Yeah, it
sounds like it hasn't tracked some core-code changes.  In particular
you might point the author at
http://git.postgresql.org/gitweb/?p=postgresql.git&a=commitdiff&h=40f908bdcdc726fc11912cd95dfd2f89603d1f37#patch10
which shows how it needs to be done in 9.0 and later.

regards, tom lane


-- 
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general

Re: [GENERAL] How to store clickmap points?

[Ben Chobot]

On Jan 8, 2013, at 2:12 AM, aasat wrote:

> Hi, 
> 
> I want to store clickmap points (X, Y and hits value) for website
> 
> I currently have table like this
> 
> CREATE TABLE clickmap (
>  page_id integer,
>  date date,
>  x smallint,
>  y smallint,
>  hits integer
> )
> 
> But this generated about 1M rows per day.
> 
> Can Postgres have better method to store this data? I also have the
> possibility to update hits value for point

Instead of storing x/y, have you considered referencing a region of pixels? The 
bigger the region, the larger your possible savings.

-- 
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general

Re: [GENERAL] query by partial timestamp

[hubert depesz lubaczewski]

On Tue, Jan 08, 2013 at 04:19:59PM -0600, Kirk Wythers wrote:
> I have a column of type TIMESTAMP, I'd like to query all records from 2011. 
> If it were text I could use a partial such as:
> 
> WHERE
>   text ~ '2011'
> 
> There must be a simple way to pull the year part out of a timestamp format. 
> Thanks in advance.

using partial checks (like extract, date_part, or even casting field to
date) will have problem with index usage.
the best way to handle it, is to write the parameters using date
arithmetic.
like:
where column >= '2011-01-01' and column < '2012-01-01'

do not be tempted to do:
where column >= '2011-01-01' and column <='2011-12-31'
which is very bad idea, and will cause data loss.

More on index usage:
http://www.depesz.com/2010/09/09/why-is-my-index-not-being-used/

Best regards,

depesz

-- 
The best thing about modern society is how easy it is to avoid contact with it.
 http://depesz.com/


-- 
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general

Re: [GENERAL] query by partial timestamp

[Michael Nolan]

It is probably not the most efficient, but I often use this syntax,
which reads better.

Select . where col_type_timestamp::date between '2011-01-01' and
'2011-12-31'

This will use a timestamp index.
--
Mike Nolan


-- 
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general

Re: [GENERAL] query by partial timestamp

[Michael Nolan]

On 1/8/13, Gavan Schneider  wrote:

> 2.  SELECT ... WHERE
>  '2011-01-01'::TIMESTAMP <= col_of_type_timestamp
>  ANDcol_of_type_timestamp <=
> '2011-12-31'::TIMESTAMP;

This won't  quite work, because '2011-12-31'::TIMESTAMP
is the same as 2011-12-31 00:00:00.0
so records timestamped later in the day on the 31st would not get selected

  SELECT ... WHERE
 '2011-01-01'::TIMESTAMP <= col_of_type_timestamp
 AND col_of_type_timestamp < '2012-01:01'::TIMESTAMP;

would get all records with a 2011 timestamp.
--
Mike Nolan


-- 
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general

[GENERAL] PostgreSQL hackfest @ Developer Conference 2013, Brno, CZ

[Honza Horak]

Hi guys,

let me announce PostgreSQL hackfest, which is held during Developer 
Conference 2013. Everyone from users, admins to hackers is welcome.


Few words about Developer Conference:
Developer Conference is an yearly conference for all Linux and JBoss 
Developers, Admins and Linux users organized by Red Hat Czech Republic, 
the Fedora and JBoss Community. 2012 DevConf had 60 talks in 3 different 
tracks and additional to that 3 more rooms with workshops and hackfests. 
With around 600 participants DevConf.cz is one of the biggest events 
about free software in the Czech Republic. You might check out photos, 
videos and blogposts about past DevConf.cz. This year the conference is 
held on February 23rd and 24th (Saturday and Sunday).


What to expect about PostgreSQL hackfest event itself:
PostgreSQL hackfest is one of the events at Developer Conference 2013, 
which is held on Sunday (2nd day of the conference) from 9am. Everyone 
around PostgreSQL project from users, admins to hackers is invited to 
meet us and talk about current PostgreSQL topics, issues or plans for 
the future. If we don't want just to talk anymore, we can hack together 
on some hot issues, so don't forget your laptop.


See the links for details:
G+ hackfest event:
  https://plus.google.com/u/0/events/cqb5364p4377vd8lp6gvmqqtk6c
DevConf web:
  http://www.devconf.cz/
DevConf schedule:
  http://developerconference2013.sched.org/

Contact person:
Honza Horak (hho...@redhat.com)

Feel free to share this announcement.

See you!

Honza


--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general

Re: [GENERAL] query by partial timestamp

[Steve Crawford]

On 01/08/2013 06:15 PM, Kirk Wythers wrote:


On Jan 8, 2013, at 6:48 PM, Tom Lane > wrote:



The OP didn't
suggest how many years his data covers, but it's quite possible that
pulling a full year's worth of data will read enough of the table that
there's no point in worrying about whether an index could be used
anyway.


There are only a few years worth of data, 2008 - 2012. However, the 
data consists of 15 min measurements and when renormalized 
(un-pivoted) is several hundred million records. It is conceivable 
that someone will want to query by month, or even hour of the day.


As another poster mentioned, you may want to consider partitioning the 
table not only for performance but also for eventual archiving/purging 
of the data.


As long as we are looking at a variety of alternatives, appropriate 
construction of partial indexes and the query *might* be of value but at 
Tom and I mentioned previously, indexes become more of a hindrance than 
a help once you start writing queries that access too much of the table 
so the planner won't use them in those cases.


Cheers,
Steve

Re: [GENERAL] query by partial timestamp

[Nathan Clayton]

On Jan 8, 2013 6:15 PM, "Kirk Wythers"  wrote:
>
>
> On Jan 8, 2013, at 6:48 PM, Tom Lane  wrote:
>
>> The OP didn't
>> suggest how many years his data covers, but it's quite possible that
>> pulling a full year's worth of data will read enough of the table that
>> there's no point in worrying about whether an index could be used
>> anyway.
>
>
> There are only a few years worth of data, 2008 - 2012. However, the data
consists of 15 min measurements and when renormalized (un-pivoted) is
several hundred million records. It is conceivable that someone will want
to query by month, or even hour of the day.

If that's the case, you may want to look at creating a date dimension and
possibly a time dimension for your data analysis (there's a good one to
start with on the PostgreSQL wiki). I would highly recommend that you take
a look at some dimensional modeling concepts (Kimball is a good place to
start).

Also, you may want to look at partitioning the data if it's several hundred
million rows.

Re: [GENERAL] Getting PLPython to work with PostgreSQL 9.2

[Adrian Klaver]

On 01/08/2013 10:38 AM, ledocf wrote:

Hi,

I have recently installed PostgreSQL 9.2 on my Windows 7 64bit computer. In
addition I have installed Python 3.3. I am new to PostgreSQL, and wish to
play with its support for Python.

I tried to run the command:

CREATE EXTENSION plpython3u

Only to receive the error:

Could not load library "C:/Program Files/PostgreSQL/9.2/lib/plpython3.dll"



Can anyone shed some light on the problems I am having? I cannot see what
environment variables need to be set (the docs I have read specify nothing).
In addition, no requirements for the supported versions of Python are
defined.


I am going to assume you are using the One-Click installer. You might 
have success asking on the installer forum:


http://forums.enterprisedb.com/forums/show/9.page



HELP!!





--
Adrian Klaver
adrian.kla...@gmail.com


--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general

Re: [GENERAL] [Solved] Corrupt indexes on slave when using pg_bulkload on master

[James Cowell]

Hi Tom,
 
I had replication enabled and it was working fine but I hadn't turned archive 
mode on yet (it was on the todo list but needed a log location).
 
I had WAL settings:
 
wal_level = hot_standby
max_wal_senders = 5
wal_keep_segments = 32
wal_buffers = 32MB
 
And checkpoint settings:
 
checkpoint_segments = 64 
checkpoint_completion_target = 0.7
 
But pg_bulkload only puts the index updates into WAL if you also have 
 
archive_mode = on
 
I guess it needs to test wal_level rather than archive mode now?  It looks like 
changes to the project have been minimal for some time, which is a shame 
because it's a very useful tool.  
 
Cheers,
 
James
 
  


 From: Tom Lane 
To: James Cowell  
Cc: "pgsql-general@postgresql.org" ; Jeff Janes 
; "w...@wdicc.com"  
Sent: Tuesday, 8 January 2013, 18:02
Subject: Re: [GENERAL] [Solved] Corrupt indexes on slave when using pg_bulkload 
on master
  
James Cowell  writes:
> I enabled archive mode (which I didn't care about before as the database only 
> holds 36 hours of data) and the indexes seem to replicate over fine.  I 
> suppose the problem here is lack of documentation, but at least the code is 
> well commented :)

> It looks like pg_bulkload works just fine with replication so long as it's 
> set up right.

Hm.  I had thought we had interlocks in there to prevent turning on
replication unless the WAL level was sufficiently high.  It sounds like
you managed to dodge that sanity check.  Could you be more specific about
what your replication configuration looks like?

            regards, tom lane


-- 
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general