Bug#439892: linux-image-2.6.21-2-amd64: random crashes/hangs on dual-core AMD64 under heavy disk write load (local and NFS)

[Graham Cobb <g+debian@cobb.uk.net>]

On Tuesday 28 August 2007 11:18, maximilian attems wrote:
> why are your reporting against that old image,
> fetch 2.6.22 from unstble.

Maks,

I run testing, not unstable.  If the problem is fixed in 2.6.22 that is good 
to know.  If you are asking me to test out the unstable kernel to report 
whether the problem still occurs I could understand that (I am certainly 
willing to try running that kernel on my production machine to test it out, 
if it works with the rest of the testing packages).  But that doesn't seem to 
be the point of your reply.

> 2.6.21 is no longer supported.

I am rather confused.  Are bug reports with testing kernels no longer welcome? 
Isn't that the point of testing?  How does the project leader work out which 
kernel will eventually ship when lenny becomes stable unless they are aware 
of all the outstanding bugs against the version currently in testing?

I could understand a reply which said "we are no longer working on 2.6.21.  It 
would be very useful if you could try 2.6.22 from unstable to see if the 
problem still occurs with that version.  Meanwhile this report will be left 
open to provide a record of outstanding bugs not fixed in the current testing 
kernel even though it will not be worked on."  But not a reply which says "do 
not report any kernel bugs unless you are running the latest unstable 
kernel".

Am I misunderstanding something?

Graham