namespaces todo list?

Jessica Frazelle me at jessfraz.com
Wed May 31 17:35:18 UTC 2017


As far as I know the only way that is done is with docker and the
default seccomp filter to block all those syscalls, you could
obviously do that with seccomp and the other runtimes as well. But
it's not a matter of just "disabling the keyring".

On Wed, May 31, 2017 at 6:32 PM, Michał Zegan
<webczat_200 at poczta.onet.pl> wrote:
> I am asking more for curiosity than because of a real need, I am just
> interested in the security of linux container technologies, and tracking
> progress. I have once heard that some linux container technologies do
> rather disable keyring access completely.
>
> W dniu 31.05.2017 o 19:26, Jessica Frazelle pisze:
>> Most container runtimes create new session keyrings per container as
>> well, idk if that helps
>>
>> On Wed, May 31, 2017 at 6:25 PM, Michał Zegan
>> <webczat_200 at poczta.onet.pl> wrote:
>>>
>>>
>>> W dniu 31.05.2017 o 19:14, Jessica Frazelle pisze:
>>>> On Wed, May 31, 2017 at 5:58 PM, Michał Zegan
>>>> <webczat_200 at poczta.onet.pl> wrote:
>>>>>
>>>>>
>>>>> W dniu 31.05.2017 o 17:23, Jessica Frazelle pisze:
>>>>>> You can catch up here[1] wrt the keyring and userns, David Howells is
>>>>>> working on more with the keyring currently[2] seems like from the set
>>>>>> of patches.
>>>>>>
>>>>>> [1] https://patchwork.kernel.org/patch/9394983/
>>>>> this patch is still in new state so not merged, hmm
>>>>
>>>> The state today is as described in that patch, which also goes over
>>>> the problems and designs. as well as the other link given which has
>>>> the more recent work.
>>>>
>>> so from what I've read in this patch, in the mailing list and even in
>>> the code it seems that the only really namespaced thing for now are
>>> persistent keyrings, and other things require consideration. Unless
>>> there is something beyond kernel/user_namespace.c that I've missed.
>>>>>> [2] https://marc.info/?l=linux-cgroups&w=2&r=1&s=David+Howells&q=b
>>>>>>
>>>>>> On Wed, May 31, 2017 at 4:17 PM, Michał Zegan
>>>>>> <webczat_200 at poczta.onet.pl> wrote:
>>>>>>>
>>>>>>>
>>>>>>> W dniu 31.05.2017 o 17:05, Jessica Frazelle pisze:
>>>>>>>>> 3 - keys, keyrings? are they namespace aware or not? I am quite lost in
>>>>>>>>> that regard, because I happen to hear conflicting statements.
>>>>>>>>
>>>>>>>> If you are using user namespaces, the keyring is namespaced.
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>> so, from which kernel version is it namespaced? and, if it really is
>>>>>>> namespaced, then does it mean the only thing not currently resolved is
>>>>>>> request_key?
>>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>
>>>>
>>>>
>>>
>>
>>
>>
>



-- 


Jessie Frazelle
4096R / D4C4 DD60 0D66 F65A 8EFC  511E 18F3 685C 0022 BFF3
pgp.mit.edu
--
To unsubscribe from this list: send the line "unsubscribe linux-security-module" in
the body of a message to majordomo at vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html



More information about the Linux-security-module-archive mailing list