In message <21252.1589106862@jrobl>, J. R. Okajima
<hooanon...@gmail.com> writes
jon bird:
That would be useful yes. I should be picking up actively looking at
this again from Monday, I'll also be able to provide some more debug as
well.
Here you are (attached).
I started picking my way through this but there seemed to be quite a lot
of differences between the 3.16 baseline and 4.19 and I don't really
know the innards of the kernel that well. So I don't think I can readily
apply this patch back very easily.
However from my earlier experiments, I had been testing out a 4.9 kernel
and was able to apply the patch to this without any difficulties.
Unfortunately due to other issues in the SELinux area I'm not able to
use this for our testing however it is good enough to verify the file
system labelling behaviour.
The good news, firstly no kernel crash and secondly it looks like the
labels are being propagated upwards correctly. My simple test program
which has a policy applied to it and writes through aufs also appears to
work (looking at the audit logs I can see no "denied" messages which is
what I was encountering before). However as you say they are quite
possibly other issues which may cause problems, mine is hardly a
definitive test.
From my perspective that is probably as far as we need to go with this,
for the purposes of my investigation getting it running on 3.16 wouldn't
tell me much more than I already know with this.
Again, thanks for your assistance with this and hopefully has
contributed something useful to the aufs development. If you need me to
try anything else out quickly whilst I have the setup in place please
let me know.
Rgs,
Jon.
--
== jon bird - software engineer
== <reply to address _may_ be invalid, real mail below>
== <reduce rsi, stop using the shift key>
== posted as: news 'at' onasticksoftware 'dot' co 'dot' uk
