Re: Bcache stuck at writeback of a key, consuming 100% CPU, not possible to detach

[Vojtech Pavlik <vojtech@suse.com>]

On Mon, Aug 31, 2015 at 05:09:28PM +0200, Emmanuel Florac wrote:
> Le Mon, 31 Aug 2015 16:49:49 +0200
> Vojtech Pavlik <vojtech@suse.cz> écrivait:
> 
> > > 
> > > The bcache_writeback stays at 100% _even_ when in writethrough mode,
> > > alas. So this looks normal. However dirty_data definitely should
> > > drop to zero...  
> > 
> > This most certainly isn't normal. The ftrace shows it's looping in a
> > loop doing nothing useful.
> > 
> 
> Yep, I've had a quick look at the code, however it's not very
> inspiring. It very much looks like there's a missing test. From your
> trace it looks like it finds a key_bad (?) -- from the struct
> bch_extent_keys_ops in bcache/extents.c -- but there is no code anywhere
> to manage this case apparently.
> 
> Ah, and I can't download your complete trace, too : twilight.ucw.cz
> isn't responding. :)

Sadly, the bcache array crashed hard earlier today (with a bug in  and the machine
doesn't boot anymore. I'm recovering it now.

-- 
Vojtech Pavlik
Director SuSE Labs