Re: DB Import Error...

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Thanks raghu ram for your immediate response,

i will try this path too along with my team and i will come back if any issues.

 

Thanks

Senthil

 


From: raghu ram [mailto:raghuchennuru@xxxxxxxxx]
Sent: Friday, April 01, 2011 8:57 PM
To: Senthil Kumar G
Cc: pgsql-admin@xxxxxxxxxxxxxx
Subject: Re: DB Import Error...

 

 

On Fri, Apr 1, 2011 at 8:34 PM, Senthil Kumar G <senthil@xxxxxxxxxx> wrote:

Hi

 

I am upgrading my staging environment from postgresql 8.2.3 to 9.0.3.

 

I was able to do successfully.

 

But, when i try to import the database which is created in 8.2.0 version to 9.0.3 version environment, i am getting following error.

 

ERROR:  constraint "xxxx” for relation "xxxx” already exist

 

Based on above error message it looks like the constraint for relation already exists.

Could you please follow below steps to upgrade from older version to new version::

Step 1: Perform the global dump on old cluster [ i.e PostgreSQL 8.2.3] using "pg_dumpall" binary of new PostgreSQL 9.0.

/opt/PostgreSQL/9.0/bin/pg_dumpall -d postgres -U postgres -p 5432 -g > /tmp/globaldump_oldpg823.sql

NOTE: It Dumps only global objects i.e roles,users and tablespaces,no databases.

Assuming old cluster running on the 5432 port number.

Step 2: Take the dump of the database in compressed format using new version of pg_dump binary.

Syntax: /opt/PostgreSQL/9.0/bin/pg_dump -Fc -v -p 5432 -f <dump file location> <database name>

nohup /opt/PostgreSQL/9.0/bin/pg_dump -Fc -v -p 5432 -f /tmp/demo.dmp demo>> /tmp/dump.log 2>> /tmp/dump.log &

-P => port number of old cluster

-Fc => compressed format

-v => verbose output

-f => location of dump file to store and dump file name

NOTE: Assuming old cluster running on the 5432 port number

Step 3: Restore the global dump on new cluster of PostgreSQL 9.0.2

/opt/PostgreSQL/9.0/bin/psql -d postgres -U postgres -p 5433 -f /tmp/globaldump_oldpg832.sql

NOTE: Assuming new cluster running on the 5433 port number

Step 4: Restore the compressed dump file using new version of pg_restore binary with parallel restore operation.

nohup /opt/PostgreSQL/9.0/bin/pg_restore -Fc -v -p 5433 -d demo -j 8 /tmp/demo.dmp >>/tmp/restore1.log 2>>/tmp/restore1.log &

NOTE: Assuming new cluster running on the 5433 port number

--Raghu Ram

 

 

 

 

when i query it was not found.

 

What could be the reason? What should i do to resolve this?

 

Appreciate your quick answer.

 

Thanks & Regards

Senthil

 

 


[Index of Archives]     [KVM ARM]     [KVM ia64]     [KVM ppc]     [Virtualization Tools]     [Spice Development]     [Libvirt]     [Libvirt Users]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite Questions]     [Linux Kernel]     [Linux SCSI]     [XFree86]

  Powered by Linux