-- Forwarded by YekFu Wong on 04/26/2001 02:27 PM
---
YekFu Wong
04/26/2001 07:26 AM
To: "Tom Lane" <[EMAIL PROTECTED]> @ INTERNET
cc:
Subject: Re: [GENERAL] Problem with pg_log file corrupted (Document link:
YekFu Wong)
I am using postgr
[EMAIL PROTECTED] writes:
> The error message appear when I tried to list all tables is as below :
> " cannot flush block 8 of pg_log to stable store "
I can't see any good reason why you'd get a failure there ... unless
maybe you are out of disk space?
Assuming you are running a 7.0.* release,
Hi,
I think the pg_log file of my postgreSQL is corrupted after a power
failure.
I can;t access to tables in my database now.
The error message appear when I tried to list all tables is as below :
" cannot flush block 8 of pg_log to stable store "
Anybody know how can I restore back the pg_log fi
>CREATE TABLE foo (test INT4) AS SELECT number FROM account;
I think you want something like:
SELECT number INTO TABLE foo FROM account ;
len morgan
---(end of broadcast)---
TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]
I'm trying to use CREATE TABLE AS under 7.03. There are no
examples in the guide, so I tried a few things:
CREATE TABLE foo (test INT4) AS SELECT number FROM account;
and
CREATE TABLE foo (test INT4) AS number FROM account;
and both of them give ERROR: parser: parse error at or near "as".
Geoff Caplan wrote:
>Please help out a Linx/Postgres newbie.
>
>I simply want to set the $PGDATA environmental variable, but can't
>figure out how. The docs assume you already know...
>
>I have tried setting it in my bash /etc/profile configuration file,
>and it shows up ok if I "e
Tom Lane wrote:
>
> Dale Walker <[EMAIL PROTECTED]> writes:
> > I use the 'hash' type as queries regarding usage will always be of the
> > form "select .. where username='xxx';"
>
> Use a btree anyway. Postgres' btree implementation is much better than
> its hash index implementation.
>
O
Dale Walker <[EMAIL PROTECTED]> writes:
> I use the 'hash' type as queries regarding usage will always be of the
> form "select .. where username='xxx';"
Use a btree anyway. Postgres' btree implementation is much better than
its hash index implementation.
> insert into sumlog
> selec
Hello,
I came up against the same (or similar) difficulty some month ago.
Then Tom Lane suggested me to write a simple sql or plpgsql function which handles the
multiple delete action.
(BTW, maybe you don't want to use the on delete cascade referential integrity in some
cases.)
Sorry, but I ca
I record our radius logs in a firly basic table, schema is as follows:
--
CREATE TABLE "history_new" (
"username" character varying(50) NOT NULL,
"time_stamp" int4 NOT NULL,
"acctstatustype" character varying(8) NOT
Joel Burton <[EMAIL PROTECTED]> writes:
> couldn't the parse tree be cached from this for each backend?
Yes, if someone wanted to work on it ...
regards, tom lane
---(end of broadcast)---
TIP 1: subscribe and unsubscribe co
On Wed, 25 Apr 2001, Tom Lane wrote:
> Joel Burton <[EMAIL PROTECTED]> writes:
> > Last night, I was doing some amateurish benchmarking and found that,
> > contrary to my (admittedly uninformed) expectation, sql functions seem
> > *slower* than plsql functions.
>
> IIRC, sql functions are re-par
On Wed, Apr 25, 2001 at 12:52:15PM -0400, Tom Lane wrote:
> Jan Ploski <[EMAIL PROTECTED]> writes:
> > Here, v_wutest is a view which returns two rows. Shouldn't the SUM
> > aggregate simply add these two values? To me, it looks like a bug.
>
> Does the view contain GROUP BY? If so, yes this is
Peter Keller <[EMAIL PROTECTED]> wrote:
>we are running PostgreSQL 7.0.2 on Solaris 2.6,/x86 compiled by gcc 2.8.1.
There are at least two components in this you might consider upgrading:
- PostgreSQL itself (to 7.0.3 or, even better, 7.1)
- gcc. Gcc 2.8.1 was a bit of a dead end in gcc developme
FYI, my book is being translated into Japanese and German. Local
publishers will be printing the translations with permission from
Addison-Wesley. I hope the translations can be made available online
like the English version.
Also, my book has been reviewed at:
http://Linuxiso.org/book
The behavior seen here seems correct although not intuitive. My guess
is that your database executables where created with locale support
enabled and your database was inited with a US locale. The sorting
logic for the english locales only sort on alpha-numeric characters.
All other charac
Hi Jan and others,
thanks for the answer, this clears up the symptom.
In trying to rewrite the rules, I'm still facing the same problem.
I'll try to simplify the rules and tables (it's emulating the OO concept as
a is the parent and b and c are inherited from a)
CREATE TABLE a (x integer PRIMAR
Jan Ploski <[EMAIL PROTECTED]> writes:
> Here, v_wutest is a view which returns two rows. Shouldn't the SUM
> aggregate simply add these two values? To me, it looks like a bug.
Does the view contain GROUP BY? If so, yes this is broken in pre-7.1
releases ...
regards, tom
On Wed, 25 Apr 2001, Ashley T. Howes wrote:
> I was wondering if PostgreSQL takes advantage of multiple CPUs when
> placed in a single Linux box. Or are multiple CPUs utilised as long
> as the Linux kernel is compiled to take advantage of them?
If you compile an SMP kernel Linux will balance
<[EMAIL PROTECTED]> writes:
> I was interested in knowing whether or not this
> is a bug, a change in the way PostgreSQL sorts, or possibly some kind of
> locale-specific misconfiguration?
There is not any (intentional) change in sorting behavior between 7.1
and earlier releases; indeed, since th
Hello Steve,
Thursday, April 19, 2001, 4:53:58 PM, you wrote:
SW> I just upgraded from 7.0.3 to 7.1 (using the RH rpms) with no
SW> major problems. Nice!
SW> However, I'm getting some strange behaviour from the init
SW> script (/etc/rc.d/init.d/postgresql). In particular, a
SW> restart whe
Hasan Mokhtari Sangchi <[EMAIL PROTECTED]> writes:
> If I want to create an index on a field by type 'timestamp'
> postgres give me following error message :
> ' can not find a default operator class for type 1296.'
> what can I do ?
Upgrade to a more recent version of Postgres.
Or, chang
On Wed, Apr 25, 2001 at 04:26:14PM +0200, Jan Ploski wrote:
> Hello,
>
> I have run into the following problem:
>
> workunit=> select * from v_wutest;
> estimate
> --
> 1650
> 100
> (2 rows)
>
> workunit=> select sum(estimate) from v_wutest;
> sum
> --
> 4950
> 10
Geoff Caplan <[EMAIL PROTECTED]> writes:
> I have tried setting it in my bash /etc/profile configuration file,
> and it shows up ok if I "echo $PGDATA" in the shell. But none of the
> postgres utilities such as "initdb" seem to be able to find it.
You probably forgot to say "export PGDATA"; witho
Lieven Van Acker wrote:
> Hi Jan and others,
>
> thanks for the answer, this clears up the symptom.
>
> In trying to rewrite the rules, I'm still facing the same problem.
> I'll try to simplify the rules and tables (it's emulating the OO concept as
> a is the parent and b and c are inherited from
Hi Andrew,
I'm sorry but I was a but sloppy in typing. The right version is
CREATE TABLE a (x integer PRIMARY KEY,y integer);
CREATE TABLE b (x integer REFERENCES a, z integer, PRIMARY KEY (x,z))
CREATE VIEW ab AS
SELECT a.x, a.y, b.z
FROM a,b
WHERE a.x=b.x;
/* this -insert- seems
will trillich <[EMAIL PROTECTED]> writes:
> "how can we cast 'varchar' to 'bpchar' or vice-versa?"
You just do it:
select bpcharfld::varchar ...
or whatever.
regards, tom lane
---(end of broadcast)---
TIP 2:
27 matches
Mail list logo
