In article <[EMAIL PROTECTED]>,
Jason Gunthorpe <[EMAIL PROTECTED]> wrote:
>
>On Thu, 26 Aug 1999, Roland Rosenfeld wrote:
>
>> The solution for this problem is to use fcntl(), because Linux 2.2.*
>> flushes the cache of a file in the moment when it is locked using
>> fcntl().
>>
>> But only fcntl() locking is not enough, because Linux 2.0.* doesn't
>> support this over NFS and then we have no locking over NFS.
>
>And linux 2.2.x with a userland server also does not support fcntl
>locking, it generates an annoying kernel message and fails with ENOLCK.
Ah but that is a configuration error. If you mount a filesystem from
a server that doesn't support locking, you are supposed to mount it
with the "nolock" option. Then the 2.2 NFS client will behave like
2.0 did - it fakes the locks locally.
However there is a bug in 2.2 that if you mount with "nolock", then
a fcntl() lock will not flush the cache anymore. Olaf Kirch
posted a patch for that on the kernel mailing list.
Mike.
--
... somehow I have a feeling the hurting hasn't even begun yet
-- Bill, "The Terrible Thunderlizards"
