Does raid5 have error handling while reading?

Does raid5 have error handling while reading?

[Xu Yuehai]

I know that if the read request of raid5 is handled by
chunk_aligned_read, there is no error handling in raid5. But, if this
read request is not aligned, that's say, raid5 need stripe_head to
deal with this request. I find in the code of 2.6.27 that there is no
error handling if this read request fail, I remember in older kernel,
some operations should be done such as compute_block. Am I right?

Re: Does raid5 have error handling while reading?

[NeilBrown]

On Fri, May 22, 2009 6:10 pm, Xu Yuehai wrote:
> I know that if the read request of raid5 is handled by
> chunk_aligned_read, there is no error handling in raid5. But, if this
> read request is not aligned, that's say, raid5 need stripe_head to
> deal with this request. I find in the code of 2.6.27 that there is no
> error handling if this read request fail, I remember in older kernel,
> some operations should be done such as compute_block. Am I right?

If chunk_aligned_read results in an error, then the request
is re-tried through the stripe cache.  I would call that "error handling".

If the cache finds that a drives is failed, the data is calculated
from the other drives, though the code is not a straight forward
as it used to be.
See fetch_block5 which sets STRIPE_OP_COMPUTE_BLK which causes
ops_run_compute5 to do the work (or to schedule the work on separate
hardware).

NeilBrown

Re: Does raid5 have error handling while reading?

[Xu Yuehai]

>
> If chunk_aligned_read results in an error, then the request
> is re-tried through the stripe cache.  I would call that "error handling".

I think you mean that error() and raid5_align_endio are invoked and
the read request is sent to conf->retry_read_aligned_list, then the
raid5d will handle it.

> If the cache finds that a drives is failed, the data is calculated
> from the other drives, though the code is not a straight forward
> as it used to be.
> See fetch_block5 which sets STRIPE_OP_COMPUTE_BLK which causes
> ops_run_compute5 to do the work (or to schedule the work on separate
> hardware).
>
I notice that the bit STRIPE_OP_COMPUTE_BLK is set if only s->uptodate
== disks - 1, however, suppose there is just a single read request, I
fail to find out why this condition is satisfied.

Yuehai

Re: Does raid5 have error handling while reading?

[NeilBrown]

On Fri, May 22, 2009 7:13 pm, Xu Yuehai wrote:
>>
>> If chunk_aligned_read results in an error, then the request
>> is re-tried through the stripe cache.  I would call that "error
>> handling".
>
> I think you mean that error() and raid5_align_endio are invoked and
> the read request is sent to conf->retry_read_aligned_list, then the
> raid5d will handle it.

Yes, that sounds right.

>
>> If the cache finds that a drives is failed, the data is calculated
>> from the other drives, though the code is not a straight forward
>> as it used to be.
>> See fetch_block5 which sets STRIPE_OP_COMPUTE_BLK which causes
>> ops_run_compute5 to do the work (or to schedule the work on separate
>> hardware).
>>
> I notice that the bit STRIPE_OP_COMPUTE_BLK is set if only s->uptodate
> == disks - 1, however, suppose there is just a single read request, I
> fail to find out why this condition is satisfied.

If there is just one read for a block on a device that has failed,
then fetch_block5 will schedule read requests (Wantread) on all
the other devices.  Once all those reads complete Uptodate will
be set, so s->update will indeed equal disks - 1, and so
COMPUTE_BLK will get set.

NeilBrown