On Mon, May 22, 2000 at 03:12:58PM -0300, Christian Robottom Reis wrote:
> On Sat, 20 May 2000, Mika Kuoppala wrote:
>
> > For all users of my read1balance patch, please
> > repatch against Ingo's 2.2.15-A0 patch
>
> Does this mean raid1readbalance-2.2.15-1 has changed again (after May 16,
> when we changed MAX_SINGLE_DISK_READS to 500)?
>
No.
raid1readbalance-2.2.15-1 fixes the bug. The change for
MAX_SINGLE_DISK_READS was just cosmetical. I will tune the
MAX_SINGLE_DISK_READS further in future, it just affects
performance.
> > There was a change that your whole mirror got frozen
> > if some of the disks were kicked out of mirror due to
> > error.
>
> I think I've seen this.. what's the change involved?
>
There was a chance of failed disk accidentally got
into voting for best candidate for next read request
--> mirror hang
Now there are more checks involving failed disks and a fall
through if there are no more then 1 working disk in array and thus
no need to go thru read balancing logic.
-- Mika
