ORA-01555
We have a process where the app team is updating a table via multiple concurrent processes. At the beginning of each process they run a full table scan against that table. In prod environment they run about 100 of these processes concurrently without issues (knock on wood)
In a test environment against a production-sized database, they ran a fewer concurrent processes and they have been failing with ORA-1555 starting couple of days ago. The test database is an exact replica of the production , but the select command that queries the whole table has been failing with ORA-1555. I have confirmed that the UNDO tablespace has the same size in both databases. The hosts themselves however are different and the test host probably has less memory and less CPU.