2015-05-01 9:52 GMT+09:00 Kohei KaiGai <kaigai@kaigai.gr.jp>:
> 2015-05-01 7:40 GMT+09:00 Alvaro Herrera <alvherre@2ndquadrant.com>:
>> Kouhei Kaigai wrote:
>>> > * Tom Lane (tgl@sss.pgh.pa.us) wrote:
>>> > > The idea of making the regression test entirely independent of the
>>> > > system's policy would presumably solve this problem, so I'd kind of
>>> > > like to see progress on that front.
>>> >
>>> > Apologies, I guess it wasn't clear, but that's what I was intending to
>>> > advocate.
>>> >
>>> OK, I'll try to design a new regression test policy that is independent
>>> from the system's policy assumption, like unconfined domain.
>>>
>>> Please give me time for this work.
>>
>> Any progress here?
>>
> Not done.
> The last version I rebuild had a trouble on user/role transition from
> unconfined_u/unconfined_r to the self defined user/role...
> So, I'm trying to keep the user/role field (that is not redefined for
> several years) but to define self domain/types (that have been
> redefined multiple times) for the regression test at this moment.
>
The second approach above works.
I defined a own privileged domain (sepgsql_regtest_superuser_t)
instead of system's unconfined_t domain.
The reason why regression test gets failed was, definition of
unconfined_t in the system default policy was changed to bypass
multi-category rules; which our regression test depends on.
So, the new sepgsql_regtest_superuser_t domain performs almost
like as unconfined_t, but restricted by multi-category policy as
traditional unconfined_t did.
It is self defined domain, so will not affected by system policy
change.
Even though the sepgsql-regtest.te still uses unconfined_u and
unconfined_r pair for selinux-user and role, it requires users to
define additional selinux-user by hand if we try to define own one.
In addition, its definition has not been changed for several years.
So, I thought it has less risk to rely on unconfined_u/unconfined_r
field unlike unconfined_t domain.
Thanks,
--
KaiGai Kohei <kaigai@kaigai.gr.jp>