Re: [dm-devel] Re: [PATCH] Always pass result and sense on request completion

[James Bottomley <James.Bottomley@HansenPartnership.com>]

On Thu, 2009-12-10 at 11:03 -0600, Mike Christie wrote:
> James Bottomley wrote:
> > On Thu, 2009-12-10 at 10:49 +0100, Hannes Reinecke wrote:
> >> Hi James,
> >>
> >> would you mind commenting on this patch?
> >> We really need this if we ever want to be able to do proper error code
> >> handling from multipath.
> > 
> > OK, not keen on the way you're setting req->errors.
> > 
> > Our big problem with FS requests is deferred or corrected errors
> > (basically we never want the FS to think we had a problem with them).  I
> > think it's currently OK because block tends to believe the returned
> > error code rather than req->errors ... I'd just hate it if that changed
> > and suddenly lots of stuff broke.
> > 
> > I think you're just looking for the sense data, so could you look at
> > that and not set req->errors?
> > 
> 
> For the specific bug Hannes is fixing we only need the sense code, so 
> that would work. If you also pass info like the host_byte bits then we 
> can do something like fail a path right away for a DID_TRANSPORT* or 
> DID_NO_CONNECT failure, but then for other errors do something else. 
> RAID could also not fail the drive on transport errors, and do something 
> different too.

So I have no trouble setting req->errors for genuine error cases (like
DID_NO_CONNECT).  I just don't want to set it for errors which are
deferred or corrected and then have something further up the stack do
the wrong thing because it thinks there was an error when, in fact,
there wasn't.

James