[Lustre-devel] [RFC] "lctl readonly" modification proposal

[Peter Braam]

Ah yes - testing requires the sync and shutdown also.  The former
unfortunately does demand an option, but the justification is now very
clear.

Peter


On 8/20/08 1:42 PM, "Mikhail Pershin" <Mikhail.Pershin at Sun.COM> wrote:

> On Wed, 20 Aug 2008 21:39:13 +0400, Peter Braam <Peter.Braam at sun.com>
> wrote:
> 
>> If I remember correctly the flush is only there to try to reduce
>> rollback.
> 
> Also to be sure that all before barrier is on disk, many test cases about
> recovery rely on the fact that operations before barrier are on disk. This
> is important with HARD failure case especially, without sync before
> barrier the operations before barrier may be lost and affect the tests.
> I agree that syncless variant of barrier is needed but we also need
> barrier with sync depending on test case.