One quick and dirty way could be to create symbolic links from files in
pgsql/data/base to an equivalent folder on your respective disks.
On Tue, 13 Jun 2000, Byron Joseph Bacaltos wrote:
> Sir,
>
> I am currently using ingres and would like to migrate to postgres.
> However the problem is that
Those are all the files you need. First, run gmake in contrib/spi and
you'll notice two more files are created:
refint.so
refint.sql
Also, take note that referential integrity on postgresql 6.5.3 works by
using triggers. So you must first CREATE FUNCTION which is detailed in
refint.sql which, in
Regarding your first suggestion, I cannot use a trigger since the actual
email contents will not be inserted in the database because of the 8K
limit on text fields. Instead, I will be keeping the messages as a
seperate file and only keep the pathname in the database.
As for your second suggestion
If there are too many requests for a particular database, is there a way
for postgresql to do load balancing across multiple servers? As far as I
can see, the only way would be to split the tables within the database
into distinct databases. Of course, this would assume there are no
dependencies b
I recommend you read Momjian's online book, very informative and provides
all the necessary information about using various kinds of id's:
http://www.postgresql.org/docs/aw_pgsql_book/index.html
As for referential integrity using foreign keys, you can add this
functionality to postgresql using tr
I think there is no maximum on the actual query length, but there is a
limit on the max row size. If I remember correctly, there is an 8k limit
to the contents of an entire row. If you need more than that, either split
your data across multiple rows are use the blob data type.
Marc
On Fri, 4 Feb
done something wrong?
Marc Tardif
like:
select email from table1, table2 where table1.oid=table2.id order by
table2.reverse_domain;
I'm not sure if keeping a secondary table is worthwhile, I will only be
executing this query about 50 times per day.
Let me know if there's a better way,
Marc Tardif
Under FreeBSD, you should be using the ports. In this case, you could
simply use the following commands in the postgresql port directory:
make
pkg_delete postgresql-6.5.2
make install
Though the procedure was simple, there was a little gotcha. You need to:
psql template1
select oid,* from pg_data
I would like to use the max row size to store data instead of using the
blob data type. From the FAQ, I understand each row must not cross 8k
boundaries so the actual data must stay below 5k.
1. Is 5k litteral, ie exactly 5120 bytes?
2. In this situation, would it be preferable to use the text d
Are rules and triggers on insert still executed when using COPY? I did a
search in chapters 8 and 13 of the programmer's guide, but I couldn't find
any mention of copy being recognised by either rules or triggers. I guess
it is assumed copy isn't the same as insert, but I would like to
double-chec
I have updated my port of postgresql on freebsd-current using the
following commands:
# make
# pkg_delete postgresql-6.5.1
# make install
Everything is working well, except for pg_dump which now returns:
pg_dump: couldn't find the template1 database. You are really hosed.
Giving up.
Yet, the tem
Here's the problem:
test=> CREATE TABLE a ( refb OID, at INT2 );
test=> CREATE TABLE b ( refa OID, bt INT2 );
test=> CREATE RULE ar AS ON UPDATE TO a.at
test-> DO UPDATE b SET b.bt = NEW.at
test-> WHERE refa = OLD.oid;
ERROR: attribute level rules currently not supported
What am I doing wron
13 matches
Mail list logo
