Datafile mistery...

Hello there,

i was just playing with user-managed bu and recovery on a Linux/9i platform to understand better the mechanism. Well... few strange things happened, strange if you take things for granted...
Do someone know why the following behaviour?

Let's start:

[Scenario: media recovery on a tablespace lost during an online backup with DML activity on it]

SQL> create table num ( id integer) tablespace users;

SQL> insert into num values (1);
SQL> insert into num values (2);
SQL> insert into num values (3);
SQL> commit;
SQL> Select * from num

   2 /

         ID

          1
          2
          3

SQL> alter tablespace users begin backup

   2 /

Tablespace altered.

SQL> alter system switch logfile;

System altered.

SQL> select name from v$datafile

   2 /

NAME

/u03/oradata/topo/sys_topo_01.ora
/u02/oradata/topo/undo_topo_01.ora
/u04/oradata/topo/users_topo_01.ora

SQL> !rm /u04/oradata/topo/users_topo_01.ora

SQL> alter tablespace users end backup

   2 /

Tablespace altered. <---- ??????

>>> Shouldn't O check for datafiles availability when
>>> the tablespace changes of status???

SQL> Select * from num

   2 /

         ID

          1
          2
          3

>>> OK: cache hit

SQL> alter system checkpoint

   2 /

System altered.

SQL> alter system switch logfile

   2 /

System altered.

>>> I was expecting O to write something on file header?!

SQL> insert into num values (5)

   2 /

1 row created.

SQL> commit

   2 /

Commit complete.

>>> cache again

SQL> alter system switch logfile

   2 /

System altered.

SQL> alter system checkpoint

   2 /

System altered.

>>> AGAIN: there is a committed tran and O
do not write to the datafile?

SQL> shutdown immediate
Database closed.
Database dismounted.
ORACLE instance shut down.

>>> File header are not touched at Shutdown?!!
>>> Confusing...

SQL> startup
ORACLE instance started.
...

>>> File header are not touched at Startup?!!
>>> More confusing...

SQL> Select * from num

   2 /
Select * from num

               *
ERROR at line 1:
ORA-00376: file 3 cannot be read at this time ORA-01110: data file 3: '/u04/oradata/topo/users_topo_01.ora'

>>> I am happy now :)

I tried to reproduce everything ona win/8i machine to see if probably there were differences between the two versions but unfortunately Win locks down the file for writing when you put the tablespace in backup mode, so i couldn't simulate a media failure.

Any idea why this behaviour? (should i ask tom? ;)

/G

PS: The only thing i was thinking is that O actually flsuh trans on the datafile when the redo log (owning the transaction) has to be overwritten, but this is not what i've been reading so far on O docs... they say that DBWr actually flushes modified blocks when a log switch occurs.

--
Remove "leva." from my email for private messages...