Hi

I have a 9i db. For one tablespace, the data of which I moved to a new tablespace in another volume, the partitions have stopped getting created.

This new tablespace had 2 datafiles with max size of 2000M each and 'Increment' set to 500M.

The size of the second datafile is showing as 1600M and the used size is showing 1502M...

Q: At what stage is a datafile extended?

Q: Could the case be that this second datafile cannot be extended and hence partitions stopped getting created.

The log shows: Inserted partition key does not map to any partition.

I might be wrong but I don't think Oracle 9i never had "automatic" partitions. You always had to create a trigger or some other maintenance procedure to create new partitions.

My first guess is that your procedure to create new partitions is no longer working.

Thanks Shammat.

Can I check any log for details on partition creation like what happened when the create partition proc ran?

Maybe the alert log but not sure what to grep from it?

Well as this is most probably a procedure/trigger/shell script you created you should know.

I don't know your system

Just wanted to add:

I've checked the alert log for 'unable to extend table......' but could not find anything.

The proc that creates partitions is running because there are 1000s of other tables which are fine.

That's why I wanted to understand whether creating partitions stop getting created if the tablespace (i.e., datafile) cannot be extended due to the increment size being too large (there is not enough free space in that datafile)?

Just looking for any place I can check what happened when the proc tried to extend the table.

>I have a 9i db. For one tablespace, the data of which I moved to a new tablespace in another volume, the partitions have stopped getting created.
>The proc that creates partitions is running because there are 1000s of other tables which are fine.

Does the proc that creates partitions know about new tablespace?

From my perspective unless or until you can post actual error code & message,
no problem really exists.

It is difficult to fix something when no clues about it being broken exist.

Yes the proc knows about the new tablespace because there is a table called 'Maintain_tablespaces' to which the new tablespace name is added.

Actually I didn't create the db. I was called in to support it. That's why some of my questions may seem naive.

There is a proc in a package called 'MAINTAIN_TABLE' which I believe creates the partitions every day.

Kindly guide me in the right direction? If you need any more info, please don't hesitate to ask.

post the code for MAINTAIN_TABLE