extent has reached it's maximum block usage

Posted by SIBSANKAR PRADHAN on 17-Dec-2015 03:01

Hi All,

Currently i am facing issue in our production database, one of the extent has reached it's maximum limit and we can't add any more extent due to it's version limitation .
In our case ,we don't need data from the affected extent but for security purpose ,we have to keep all the records in that extent so, we can't delete any data to reduce the size .
No application level data archiving policy configured in our database .

Can you please suggest me urgently about the below suggestion -
1) Create a new Database - not possible immediately
2) Upgrade progress version - not possible immediately
3) Create a new data area and update the same in the .st file and load a new df into the database , so that new records can be stored in the new data area and old records
also can be recoverable from the old data area by loading the old df ,if needed .
Here i am planning to apply this solution in production

Version and DB details:
Progress version - 9.1E
Database Size - 800GB
Records per block - 128
DB Block size - 8k

Please advise me regarding our ideas and also any other idea that can be feasible in this scenario.

All Replies

Posted by James Palmer on 17-Dec-2015 04:10

Why are you on 9.1E? Why is an upgrade not possible? You really should upgrade to 10.2B08 at the very least, and urgently.

On the topic of your problem, how many tables are in the storage area that is full?

Posted by SIBSANKAR PRADHAN on 17-Dec-2015 04:20

The project will start migrating to another technology  with in next 4-5 months ,so client has not agreed to upgrade progress version

Only one table is defined in that area and it's growth size is actually very high (1GB /day ) and about to exceed  it's limit  .  

Posted by James Palmer on 17-Dec-2015 05:00

Others may have more technical solutions to this, but I see 2 options:

1) Upgrade sooner

2) Archive data, even if it's temporary until an upgrade has occurred.

Posted by Libor Laubacher on 17-Dec-2015 05:31

Dump and load into restructured database with multiple storage areas. Ie spread the data even more.

>>

Create a new data area and update the same in the .st file and load a new df into the database , so that new records can be stored in the new data area and old records also can be recoverable from the old data area by loading the old df ,if needed

<<

For that to work, you would need to rename the affected tables to something else if you plan to do this on your current database, ie you can't have same table (name) in 2 storage areas. Might be better to create a new database with just those tables if you want to start fresh and have them both connected, but then your application would need to refer to those tables using ldbname.table.

This thread is closed