Linus Torvalds wrote:
> ...
>
> Because the low-level filesystems _have_ already re-tried. So there's no
> point in the MD device doing the same thing. Once a low-level device has
> an error, we've done all the retries it's sane to do (sometimes a lot
> more), and MD retrying would only make erro
Veritas Volume Manager has a "virtual" device driver (vxio) which handles the
intermediary step between the actual device driver and the higher level.
This allows for some advanced RAID possibilities - if a write fails to
complete, the ioctl returns a value of -1... That way you can execute a
Veri
> Umm. Isn't RAID implemented as the md device? That implies that it is
> responsible for some kind of error management. Bluntly, the file systems
> don't declare a file system kaput until they've retried the critical
> I/O operations. Why should RAID5 be any less tolerant?
File systems give up t
> any data, but under normal default drive setup the sector will not be
> reallocated. If testing the failing sector is too much effort, a
> simple overwrite with the corrected data, at worst, improves the
> chances of the drive firmware being able to reallocate the sector.
> This works just f
Alan Cox wrote:
>
> > > 1) Read and write errors should be retried at least once before kicking
> > >the drive out of the array.
> >
> > This doesn't seem unreasonable on the face of it.
>
> Device level retries are the job of the device level driver
>
> > > 2) On more persistent read error
On Wed, 21 Mar 2001, Max TenEyck Woodbury wrote:
>
> Umm. Isn't RAID implemented as the md device? That implies that it is
> responsible for some kind of error management. Bluntly, the file systems
> don't declare a file system kaput until they've retried the critical
> I/O operations. Why shoul
Alan Cox wrote:
>
>>> 1) Read and write errors should be retried at least once before kicking
>>>the drive out of the array.
>>
>> This doesn't seem unreasonable on the face of it.
>
> Device level retries are the job of the device level driver
Umm. Isn't RAID implemented as the md device?
> > 1) Read and write errors should be retried at least once before kicking
> >the drive out of the array.
>
> This doesn't seem unreasonable on the face of it.
Device level retries are the job of the device level driver
> > 2) On more persistent read errors, the failed block (or whatever u
(I've taken Alan and Linus off the Cc list. I'm sure they have plenty
to read, and may even be on linux-raid anyway).
On Thursday March 15, [EMAIL PROTECTED] wrote:
> I'm not too happy with the linux RAID5 implementation. In my
> opinion, a number of changes need to be made, but I'm not sure how
Max TenEyck Woodbury wrote:
>
> I'm not too happy with the linux RAID5 implementation. In my
> opinion, a number of changes need to be made, but I'm not sure how to
> make them or get them accepted into the official distribution if I did
> make the changes.
>
> The changes I think should be made
10 matches
Mail list logo
