To re-enable access to a down table, you can submit an appropriate ALTER TABLE … RESET DOWN SQL request. This removes the down status flag from all AMPs in the system and makes the table available for all DML requests.
When you reset the down status flag for a table by submitting an ALTER TABLE … RESET DOWN request, Teradata Database removes the down status flag in the table header pertaining to all the subtables from all the AMPs. The system also deletes the down regions from the table header on each AMP except for regions that were skipped during rollback and recovery processing.
To avoid potential transaction aborts due to file system faults, you should submit your ALTER TABLE … RESET DOWN requests only after you repair the intermittent hardware errors that caused the table to be flagged as down in the first place. If the intermittent hardware issues that caused data blocks to be marked bad are not resolved beforehand, then a DML request can attempt to access them, triggering another round of file system aborts and flagging regions down. Any DML requests that access rows in data blocks that do not have problems continue to execute successfully.
- Run the Table Rebuild utility. For more information, see Teradata Vantage™ - Database Utilities , B035-1102 .
- Submit a fast path DELETE ALL request, which effectively ignores the current rows in the table. See Teradata Vantage™ SQL Data Manipulation Language , B035-1146 .
If the down table is later rebuilt or restored from an earlier backup, Teradata Database resets the down status flag for the various subtables and moves their rows to unmarked data blocks and cylinders from the fallback copies.