[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <f9af66c8-955d-0788-3e70-3a1cd1e96267@linux.ibm.com>
Date: Wed, 26 Jan 2022 16:54:17 -0500
From: Stefan Berger <stefanb@...ux.ibm.com>
To: Christian Brauner <brauner@...nel.org>,
Stefan Berger <stefanb@...ux.vnet.ibm.com>
Cc: linux-integrity@...r.kernel.org, zohar@...ux.ibm.com,
serge@...lyn.com, christian.brauner@...ntu.com,
containers@...ts.linux.dev, dmitry.kasatkin@...il.com,
ebiederm@...ssion.com, krzysztof.struczynski@...wei.com,
roberto.sassu@...wei.com, mpeters@...hat.com, lhinds@...hat.com,
lsturman@...hat.com, puiterwi@...hat.com, jejb@...ux.ibm.com,
jamjoom@...ibm.com, linux-kernel@...r.kernel.org,
paul@...l-moore.com, rgb@...hat.com,
linux-security-module@...r.kernel.org, jmorris@...ei.org
Subject: Re: [PATCH v9 11/23] ima: Move ima_lsm_policy_notifier into
ima_namespace
On 1/26/22 08:05, Christian Brauner wrote:
> On Tue, Jan 25, 2022 at 05:46:33PM -0500, Stefan Berger wrote:
>> From: Stefan Berger <stefanb@...ux.ibm.com>
>>
>> Move the ima_lsm_policy_notifier into the ima_namespace. Each IMA
>> namespace can now register its own LSM policy change notifier callback.
>> The policy change notifier for the init_ima_ns still remains in init_ima()
>> and therefore handle the registration of the callback for all other
>> namespaces in init_ima_namespace().
>>
>> Signed-off-by: Stefan Berger <stefanb@...ux.ibm.com>
>> ---
> I'd double-check that this cannot be used to cause rcu stalls when a lot
> of ima namespace with a lot of rules are used leading to a dos situation
> during LSM policy update. The good thing at least is that an LSM policy
> update can only be triggered for selinux for the whole system.
I just ran a test with up to 1920 IMA-namespaces each with 2 audit rules
with the vmtools_exec_t label. Disabling of the vmtools SELinux module
caused the rules to disappear in all IMA namespaces, as expected.
However, it also added many kernel log lines 'ima: rule for LSM
'vmtools_exec_t' is undefined' to the kernel log that should probably be
suppressed for ns != &init_ima_ns. Nothing bad happened otherwise. Also
re-enabling the vmtools module didn't cause any kernel errors. So I
think we should be fine.
https://github.com/stefanberger/ima-namespaces-tests/tree/master/audit-many-2
Powered by blists - more mailing lists