I have a table (5 columns) with approximately 670 million rows. It has had an index (unique) on an inet column from the beginning. Today I added a primary key constraint based on the same column thinking that since it already had an index, this would be a relatively quick operation. That does not appear to be case. It has gone into a "not responding" status for an hour or so now. As a point of reference, I'm using 9.3 on a 64 bit Windows Server 2008 (32 GB ram) and inserts so far have taken 6 1/2 - 7 minutes for each batch of 16.7 million rows.
Other than not creating the primary key at the beginning, did I do anything wrong? and can I reasonably expect the current operation to finish?
It will finish but it is checking that not only is it unique, but NOT NULL. You may want to cancel it and try:
select count(*) from foo where bar IS NULL;
If you have NULL in the column, you don't have a primary key
Joshua D. Drake
Joey
-- Command Prompt, Inc. - http://www.commandprompt.com/509-416-6579 PostgreSQL Support, Training, Professional Services and Development High Availability, Oracle Conversion, Postgres-XC, @cmdpromptinc For my dreams of your image that blossoms a rose in the deeps of my heart. - W.B. Yeats
