If I try to delete a record from my table which is giving following error
Warning: PostgreSQL query failed: ERROR: pg_atoi: zero-length string in
/xxx/database.inc on line 73
What might be the problem.
Regards,
Sreedhar Bhaskararaju
---(end of broadcast)---
Hi-
I tried to post this to the performance list, but that list seems to have a
problem at the moment. I think the question fits "admin" as well:
I have a table- called "event" with a field event_date_time that is indexed.
There are 1,700,000 rows in the table and 92,000 distinct values of
event_
Oops- forgot to give the version on that last question:
I'm running version 7.3.2 on a Debian Linux platform.
-NF
-
Nick Fankhauser
[EMAIL PROTECTED] Phone 1.765.965.7363 Fax 1.765.962.9788
doxpop - Court records at your
On Mon, 22 Sep 2003, Jodi Kanter wrote:
> I have inherited a database that was originally created in postgres
> version 6.? There were no steps taken at that time to implement
> constraints to ensure referential integrity. There are no FK
> constraints, PK constraints, etc..
> Most of what curr
On Mon, 22 Sep 2003, Hornyak Laszlo wrote:
> Hi all!
>
> We have a database on postgreSQL 6.2 and it is extremely slow, so we
> started vacuum on it. I know it locks the tables, so clients can not use
> it until the process is finished, but it is extremely slow on a 1.800.000
> record table and w
Since I am a novice can you explain more about this version mismatch?
From: Bruce Momjian <[EMAIL PROTECTED]>
To: Priya G <[EMAIL PROTECTED]>
CC: [EMAIL PROTECTED]
Subject: Re: [ADMIN] help needed!!!
Date: Mon, 22 Sep 2003 14:46:27 -0400 (EDT)
Priya G wrote:
[ text/html is unsupported, treating li
"Nick Fankhauser" <[EMAIL PROTECTED]> writes:
> So the question is- how can I get a better estimate of n_distinct from
> analyze?
> If I alter the stats target as high as it will go, I get closer, but it
> still shows the index to be about 1/2 as selective as it actually is:
AFAIK, estimating numb
> AFAIK, estimating number of distinct values from a small sample is
> inherently an ill-conditioned problem.
If I had been getting estimates all over the map, I'd have been a bit more
unconcerned, but what I'm seeing is a very consistent number that also
increases and tends to be more consistent
Sure, basically, modern versions of the client library can talk to older
versions of the database backend up to a point (i.e. the 7.4 library in
beta can go back to 6.5 or 6.4 or something like that. Support for 6.2 or
thereabouts was recently dropped.)
However, going the other way, old connec
"Nick Fankhauser" <[EMAIL PROTECTED]> writes:
> when I do a \d on pg_stats, it appears that pg_statistic.starelid matches up
> with pg_class.oid, but apparently this is not the case.
It certainly should be the case. starelid matches to pg_class.oid and
staattnum matches to pg_attribute.attnum.
> It certainly should be the case. starelid matches to pg_class.oid and
> staattnum matches to pg_attribute.attnum.
My problem was that I was looking up "event_date_time" in pg_class.relname
(and finding it), but the oid matched nothing. when I looked for 'event' in
pg_class & 'event_date_time'
Hi Jodie,
This snippet is generated code from my client side data dictionary for
postgresql.
The sequence described is used with nextval() to obtain autoincrement
key values
for the primary key. Thus primary key definition does create an index as
do for
"Nick Fankhauser" <[EMAIL PROTECTED]> writes:
> So... I updated stadistinct to the actual value of 92,000 as you suggested
> and as you predicted, it did not change the plan a bit.
Just out of curiosity, what happens if you make it bigger than 92k?
Does a value 10x or 100x reality change the plan?
> Just out of curiosity, what happens if you make it bigger than 92k?
> Does a value 10x or 100x reality change the plan?
Neither one makes a change- perhaps something else is at work here- my
understanding of the finer points of query plans is shaky- Here is the query
and the plan I'm getting:
I wish to have a Java application connect to PostgreSQL via
SSL. I know how to set up postgres to accept SSL connections, but I do
not know how to use the JDBC driver to establish the connection. Can
anyone point me in the right direction? Thanks.
Rob Abernethy
"Nick Fankhauser" <[EMAIL PROTECTED]> writes:
> Neither one makes a change- perhaps something else is at work here- my
> understanding of the finer points of query plans is shaky- Here is the query
> and the plan I'm getting:
Well, one thing that's not a fine point is that if you are complaining
a
scott.marlowe writes:
> Sure, basically, modern versions of the client library can talk to older
> versions of the database backend up to a point (i.e. the 7.4 library in
> beta can go back to 6.5 or 6.4 or something like that. Support for 6.2 or
> thereabouts was recently dropped.)
>
> However,
Hello,
User authorization is a difficult task. Requiring passwords ist
annoying, Kerberos is complicated, ident is not secure.
What about leaving it to the administrator of a system to provide a
function (either a C function in a shared lib or an executable) which
gets information passed and deci
On Tue, Sep 23, 2003 at 10:52:27AM -0600, scott.marlowe wrote:
> On Mon, 22 Sep 2003, Jodi Kanter wrote:
>
> > I have inherited a database that was originally created in postgres
> > version 6.? There were no steps taken at that time to implement
> > constraints to ensure referential integrity.
Oops- forgot to give the version on that last question:
I'm running version 7.3.2 on a Debian Linux platform.
-NF
-
Nick Fankhauser
[EMAIL PROTECTED] Phone 1.765.965.7363 Fax 1.765.962.9788
doxpop - Court records at your
Hi Newsgroup!
I can't believe I have found a good solution here. I want to return the
last value of an array with an SQL statement, while I do not know, how many
items the array holds (u can never be sure of that since u can redim an
array in PostgreSQL whenever u like..).
That is what i came
> I can't believe I have found a good solution here. I want to return
> the last value of an array with an SQL statement...
> That is what i came up with:
>
> select my_array_field[substr(array_dims(my_array_field), 4, 1)]
> from my_table;
>
> But this only works, as long as the upper boundary is f
scott.marlowe wrote:
> On Tue, 16 Sep 2003, John Allgood wrote:
>
> > Hello
> >
> > I am looking for information on what operating systems and
> > filesystems people are running postgresql on. I have read so much on this I
> > decided to get some input from other people. I at first was leani
Holger Jakobs <[EMAIL PROTECTED]> writes:
> What about leaving it to the administrator of a system to provide a
> function (either a C function in a shared lib or an executable) which
> gets information passed and decides whether the access will be granted?
I think this is exactly what PAM is for.
(B
(B
(BCould someone please tell me where I could
(Bfind info on exporting excel data into postgresql, or how to do it.
(BThanks!
(B
On Wed, Sep 24, 2003 at 15:13:40 +0900,
Lee Weaver <[EMAIL PROTECTED]> wrote:
> Could someone please tell me where I could find info on exporting excel data
> into postgresql, or how to do it. Thanks!
If you export your excel data into tab separated data files (one for
each table), then it will
(B
(B
(B
(BLee Weaver írta:
(B
(B
(B
(B
(B Could someone please tell me
(Bwhere I could find info on exporting excel data into postgresql, or how
(Bto do it. Thanks!
(B
(B
(BTry to save your data from excel in csv format and then use a COPY
(Bto load it to a previously cr
Dear all,
I am running a PostgreSQL 7.3.2 on a FreeBSD. A Tomcat server
connects to it locally by JDBC.
I created a test module which sends a http request to the Tomcat
server to store some data on DB. The test module sends a http
request 5000 times continusly. After sending 5000 requests, a lot
shreedhar wrote:
>
> I am using PHP4 as client side.
>
[snip]
> Subject: Re: [ADMIN] Fatal error: Allowed memory size of 8388608 bytes
> exhausted
>
this is a php appache message. either
A) you have a phpscript that allocates memory in a infinate loop
B) your php script has become so big, or its
Looks like line 73 of your /xxx/database.inc PHP script is submitting a query
formated with a zero length string instead of a number. Look for the variable
being formated into the query string being undefined thereby leading to the
somethat that looks like:
delete from aaa where id =
instead of
On Tue, 23 Sep 2003, shreedhar wrote:
> If I try to delete a record from my table which is giving following error
>
>
> Warning: PostgreSQL query failed: ERROR: pg_atoi: zero-length string in
> /xxx/database.inc on line 73
>
> What might be the problem.
Looks like line 73 of your /xxx/database
If I try to insert I got the following error as
Fatal error: Allowed memory size of 8388608 bytes exhausted (tried to
allocate 79 bytes)
Where to set that.
Regards,
Sreedhar Bhaskararaju
---(end of broadcast)---
TIP 1: subscribe and unsubscribe
"shreedhar" <[EMAIL PROTECTED]> writes:
> If I try to insert I got the following error as
> Fatal error: Allowed memory size of 8388608 bytes exhausted (tried to
> allocate 79 bytes)
There is not, and never has been, any error message of that wording in
the PG backend. You must be looking at some
I am using PHP4 as client side.
- Original Message -
From: "Tom Lane" <[EMAIL PROTECTED]>
To: "shreedhar" <[EMAIL PROTECTED]>
Cc: "Postgre General" <[EMAIL PROTECTED]>; "Postgre Admin"
<[EMAIL PROTECTED]>
Sent: Wednesday, September 24, 2003 10:44 AM
Subject: Re: [ADMIN] Fatal error: Allowe
34 matches
Mail list logo
