Re: Restore postgres database problem

Yes it returning carriage returns. I am pasting below the output to the
command u send to me

*******Linux Prompt**********************
[root@localhost inpreet]# head postgres.sql | cat -v
--^M
-- PostgreSQL database cluster dump^M
--^M
^M
\connect "template1"^M
^M
--^M
-- Users^M
--^M
^M
***************************************

Now what should I do?

Regards
Inpreet Singh
----- Original Message -----
From: "Michael Fuhr" <mike@fuhr.org>
To: "Inpreet Singh" <Singh.Inpreet@netsity.com>
Cc: <pgsql-admin@postgresql.org>
Sent: Friday, January 28, 2005 10:34 AM
Subject: Re: [ADMIN] Restore postgres database problem


> On Thu, Jan 27, 2005 at 11:03:43AM +0530, Inpreet Singh wrote:
>
> > *Taken Dump :*   /usr/bin/pg_dumpall > /home/inpreet/postgres.sql
/*which
> > was taken at client site*/
> >
> > *Restoring Dump :*   su - postgres /*This process I am doing at local
site
> > so thats why psql and dumpall are in diff directories*/
> >  /usr/local/pgsql/bin/psql -d template1 < /home/inpreet/postgres.sql
>
> Is the local site running the same version of PostgreSQL or a later
> version than the client site?  If you try to restore into an earlier
> version then you could have problems.
>
> How are you transferring the file from the client site to the local
> site?  Is it possible that the file was modified in transit?  I
> wonder if carriage returns have been added -- do you see ^M at the
> ends of lines when you run the following command?
>
> head postgres.sql | cat -v
>
> > Sir I have some table values which are of type timestamp and those
columns
> > have values like
> > ""2004-10-09 10:03:36.689993"". So is it possible that when I am
restoring
> > values this columns are creating all sort of trouble??? as timestamp
field
> > means date time at the time of insertion of record and I am trying to
insert
> > earlier date time.
>
> The "timestamp" type holds a date and time; PostgreSQL shouldn't
> care about the value as long as it's syntactically correct (although
> it's possible that a trigger or constraint might care).
>
> --
> Michael Fuhr
> http://www.fuhr.org/~mfuhr/
>