Re: Re: General question

"Also sprach Stephan von Krawczynski:"
> On Mon, 6 Jun 2005 22:06:17 +0200 (MET DST)
> "Peter T. Breuer" <[email protected]> wrote:
> 
> > [...]
> > > This is from nbd-2.4.32pre.tgz I downloaded around 6th April 2005.
> > 
> > Looks fine - seems you just have the 2.4 code, and anyway it seems
> > maybe I left it up to the admin to set show_errs.  But I think it is
> > probably OK to set it on going into a raid array when its a FR1 array.
> > And since only FR1 sends the two ioctls above, it should be fine to add
> > the appropriate code (sketched above) there.
> > 
> > I've done it now in my copy.
> 
> Can I download this version somewhere? Just to be sure we are talking of the
> same thing in the future. I just downloaded the version from 6th June, and it
> is not in there.

Well, it wouldn't be - I added it into my copy. I'm waiting for you to
tell me it doesn't kill your grandmother before making the change
public!

> One thing that looks almost like a bug is the fact, that after removal of a
> failing unit the show_err state gets dropped in the version I have.

Well, it probably would - but I don't think you should worry too much.
If you make the small addition I specified, when the device comes back
up it will be re-added to the array, which I THINK will turn show_errs
back on.

Actually, no, I don't think the current code should turn off show_errs, but
I guess we can afford to pass on looking at that for the moment. 

> So if the
> device is re-enabled (because network comes back) it won't work a second time,
> but hang _then_. You would have to set the flag manually again after such a
> failure, but you don't know when...
> 
> Basically I think the approach to set it during this ioctl looks best.


Tell me how your grandmother feels about it.

Peter