Re: Async I/O is NOT supppressed for filesystem datafiles in 8.1.5?

Can you clarify your problem:
You say

    >You're misunderstanding.  I'm very well versed in
    >the difference between kernel aio and filesystem
    >aio on Solaris.  In my case, DBWR is using the
    >latter against my datafiles.

But you also say

    >So, why is my instance still trying to use it?

Are you complaining because your system IS using aync I/O, or because your system is TRYING to use Async I/O ?

A couple of ideas that may help:

1. The manuals are not always correct
2. The analysts who answer questions on Metalink often do so after reading the manuals
3. You have the system in front of you - if you can see that the system is contradicting the manual and the analyst, then the manual is wrong.
4. If you want confirmation of your interpretation of what you are seeing, then it should be pretty easy to post the truss output that has convinced you.

Thank you for pointing out my error regarding NFS, and the NetApps, and more thanks to Paul Fagan for posting a useful URL.

--
Jonathan Lewis
http://www.jlcomp.demon.co.uk

Now running 3-day intensive seminars
http://www.jlcomp.demon.co.uk/seminar.html

Host to The Co-Operative Oracle Users' FAQ
http://www.jlcomp.demon.co.uk/faq/ind_faq.html

Author of:
Practical Oracle 8i: Building Efficient Databases


El Toro wrote in message <2e1cd2b.0112280628.63eb2f9f_at_posting.google.com>...


>You're misunderstanding.  I'm very well versed in

>the difference between kernel aio and filesystem

>aio on Solaris.  In my case, DBWR is using the

>latter against my datafiles.

>

>However, Oracle claims that the oracle process will

>determine whether filesystem files are in use, and

>if so will turn OFF disk_asynch_io -- even if you

>have it set to true (check Metalink DocID 214480.999

>for details).

>

>So, why is my instance still trying to use it?

>

>

>Martin Haltmayer <Martin.Haltmayer_at_d2mail.de> wrote in message

news:<3C2C4693.B6D70062_at_d2mail.de>...


>> Please show the truss output. Oracle by itself does not prevent kaio

calls. But


>> in Solaris they return with an ENOTSUP exception and Oracle reverts to

>> pwrite/pread calls. So Oracle tries whatever you tell it in init.ora. If

you put


>> disk_asynch_io = true there it will try kaio calls. If they fail it will

go back


>> to pread/pwrite.

>>

>> On Solaris it is safe to request kaio on filesystems but it is slow

because you


>> add the overhead of exception handling *after each kaio call*.

>>

>> Regards,

>>

>> Martin