Wenn es nicht an den Festplatten gelegen hat, wirst du den Fehler nicht finden. Wir hatten vor geraumer Zeit bei vier Datenbanken das gleiche Phänomen. Die Lösung des Problems hat uns nicht interessiert sondern der grund für die Fehlermeldung. Nach langem Hin und Her mit dem Support von IBM (Ireland) haben wir dann diese eMail erhalten:
Please provide me with a list of steps so that I can attempt to reproduce
the corruption.
The residual corruption can tell us what is corrupt in the database, but not what series of events brought the database to that state. Database corruption is when some parts or all of the database objects are unreadable or the integrity is questionable. A corrupt state can occur in several places. These related to bad tables, notes, folders, bitmaps and so on.
Questionable integrity occurs when a database header or index has entries that appear to be invalid. To some extent, an unknown state could occur that prevents any Domino task from performing as it should.
Causes of Database Corruption
Improper operation procedures (for example, improper shutdown)
Operating system or network problems
Memory overlay
Hardware failures
In any event, these environmental hazards cause abrupt termination that
prevents a database from writing to a disk or being closed properly after
it was accessed. An unknown state occurs when the database is not closed
properly, for example, a server crashes or a database file is copied at the
operating system level while open.
Read-Only Corruption
When the Domino server detects a corrupt database, it will mark it as read
only, and when the
database is opened, the Domino server shows error messages similar to the
following.
"NSFDbOpen: File 'C:\louts\notes\data\test.nsf' is CORRUPT - Now
Read-Only!"
-or-
"Database is corrupt -- Cannot allocate space"
Undetected Corrupted Database
This happens only when any of the server tasks are unable to recognize the
integrity of the database table, index, view and so on. The effects may
cause a server to crash. You should search the Notes log/console output for
the last couple of crashes and compare similarities of any particular
databases that AMgr, Indexer or any other task was working on before the
crash.
I'm afraid I can't help you. Let me know how you would
like to progress this PMR.
MOD
