Re: [GENERAL] More information

I've seen this behavior in 6.3.2 also.  It went away in 6.4.

Steve

On Thu, 11 Mar 1999, M Simms wrote:

> Thanks for the response. However, this is only part of the problem, and this
> will not fix it.
>
> I have done a vacuum, which seems to succede, but does not fix the error.
>
> If I destroy the database, and recreate it, I get amcreate: relationship
> already exists *as I create the tables for the first time in the
> database* The database and tables are of the same name as the old
> ones, of course, but the old ones *should be gone* but they arent.
>
> Example:
> *************************
> $destroydb mydb
> $destroydb mydb
> ERROR:  destroydb: database mydb does not exist.
> destroydb: database destroy failed on mydb.
> $createdb mydb
> $echo create table beep \( test int \)\; | psql mydb
> create table beep ( test int );
> ERROR:  amcreate: beep relation already exists
> EOF
> *************************
>
> This means no matter what i do, I cannot even use the same database
> name again!
>
> Any further ideas would be *greatly* appreciated. Surely there has to
> be something like a fsck for the database where it just does an
> integrity check and asks you if something is an error, should it be
> removed/fixed... ?
>
> Thanks again
>
>                     M Simms
>
>
> >
> > Try the vacuum command and see if that cleans up the database.
> > Then, if that fails, instead of dumping the entire db, use
> > the "pg_dump -t table dbname > dbname.table.out" to dump
> > each table that is still in good shape. Sorry I can't help
> > you with what the root cause is, but the above may get you
> > out of the jam.
> >
> > Thomas
> >
> > M Simms wrote:
> > >
> > > Okee - more information on my problem.
> > >
> > > Postgresql version 6.3.2 on redhat linux 5.2
> > >
> > > Here is a cut&paste from what happens
> > > ************************
> > > $ psql mydb
> > > Welcome to the POSTGRESQL interactive sql monitor:
> > >   Please read the file COPYRIGHT for copyright terms of POSTGRESQL
> > >
> > >    type \? for help on slash commands
> > >    type \q to quit
> > >    type \g or terminate with semicolon to execute query
> > >  You are currently connected to the database: mydb
> > >
> > > mydb=>\d
> > >
> > > Database    = mydb
> > >  +------------------+----------------------------------+----------+
> > >  |  Owner           |             Relation             |   Type   |
> > >  +------------------+----------------------------------+----------+
> > >  | myuid            | beep                             | table    |
> > >  | myuid            | history                          | table    |
> > >  | myuid            | history_seq                      | sequence |
> > >  | myuid            | historyb                         | table    |
> > >  | myuid            | other                            | table    |
> > >  | myuid            | record                           | table    |
> > >  | myuid            | recordb                          | table    |
> > >  +------------------+----------------------------------+----------+
> > > mydb=> select * from history_seq;
> > > sequence_name|last_value|increment_by|max_value|min_value|cache_value|is_cycled|is_called
> > > -------------+----------+------------+---------+---------+-----------+---------+---------
> > > (0 rows)
> > > mydb=> drop sequence history_seq;
> > > ERROR:  DeletePgTypeTuple: history_seq type nonexistent
> > > mydb=>
> > > ************************
> > >
> > > I have no idea what to do about this, cos I need to update my database
> > > but I cant do it with this happening, cos when I dump the database I
> > > get
> > >
> > > ************************
> > >
> > > $ pg_dump mydb > dump.out
> > > dumpSequence(history_seq): 0 (!= 1) tuples returned by SELECT
> > >
> > > ************************
> > >
> > > I need some solution to this if anyone knows. Surely a database this
> > > complex HAS to have some sanity-checking routines and database
> > > restoration. The documentation helpfully says that this section is
> > > missing, however. I have based a LARGE project on this, and so far, I
> > > cant start this project till this problem is fixed.
> > >
> > > Thanx
> > >
> > >                                 M Simms
> >
>
>
>