I'm on a client now in a very strange situation...
a couple of days ago, the client told me that oracle was consuming a lot of disk space, until the server hd came totally full... so the oracle stops working. The TI guy was deleting other old files that hasn't relevance for oracle, just to release disk space, just to keep oracle working... so I could arrive here few minutes ago.
On trace, just have one simple line, that set to a trace file...
Mon May 11 13:35:43 2009
Errors in file d:\oracle\admin\orcl\udump\orcl_ora_1436.trc:
ORA-00600: internal error code, arguments: , , [0x15CA4B70], , , , , 
ORA-00600: internal error code, arguments: , [0x93D0904], , , , , , 
I went to that, to see what kind of error happened. This file had 90gb and was growing up very fast... There was other in BDUMP, same size, same growing up.
So I tried to delete both.. and cannot do it, because oracle was sending more and more information do them...
Then I tried to stop oracle service and delete it... cannot to... "The computer cannot stop the local service on a remote computer (something like that)"
So I restart the server, deleted the files, and wait for a new file to open it and see the error.
Now oracle works very fine. No more big trace files on both directories... and no more message error on alter... just the alter database open
Someone could know what happened here?
I'm affraid that after I leave the client the same problem occurs again...