Hi!

I have a Win XP Computer and sequoia 2.10.6, using a Sequoia Radib-01
setup with Postgres 8.2 as backends. The backends are located on the
same physical computer, but under different names (DB_01, DB_02). The

My database (DB_01) consists of two tables as follows:

CREATE TABLE demo
(
 x integer
)
WITHOUT OIDS;
ALTER TABLE demo OWNER TO postgres;

and

CREATE TABLE test
(
 id integer NOT NULL,
 name character varying(255),
 CONSTRAINT test_primarykey PRIMARY KEY (id)
)
WITHOUT OIDS;
ALTER TABLE test OWNER TO postgres;
GRANT ALL ON TABLE test TO postgres;

I use altered demo scripts to start one controller with my two
backends, and initialize it.

During initialization, I backup DB_01, using Octopus, placing the
backup in /tmp. Then I and enable that backend. All is good.
The first problem occurs when I try to restore DB_02 using the backup
dump I just created. It seems that the restore app looks for my backup
in c:/tmp/tmp/..., where it does not exist. I work around this problem
by copying my backup to c:/tmp/tmp, while at the same time leaving it
at its original location. After this I am able to restore the DB_02.
No error messages occur, but looking at the table definitions in the
postgres admin tool the DB_02 tables (created by the restore) look as
follows:

CREATE TABLE demo
(
 x character varying
)
WITHOUT OIDS;
ALTER TABLE demo OWNER TO postgres;

CREATE TABLE test
(
 id character varying NOT NULL,
 name character varying(255)
)
WITHOUT OIDS;
ALTER TABLE test OWNER TO postgres;

As yoiu might suspect, when I try to enable DB_02, I get an error
stating that I have a column definition mismatch in the demo table
between the two databases. The database DB_02 is still, however,
enabled, dispite the error message that I get.

So, any suggestions as to what might be wrong?

Thanks,

Thom Birkeland
Diamo AB

_______________________________________________
Sequoia mailing list
[email protected]
https://forge.continuent.org/mailman/listinfo/sequoia

Reply via email to