Re: [dm-devel] [PATCH 0/7] device mapper target measurements using IMA

From: Thore Sommer <public@thson.de>
To: tusharsu@linux.microsoft.com
Cc: agk@redhat.com, dm-devel@redhat.com,
	linux-integrity@vger.kernel.org, nramas@linux.microsoft.com,
	public@thson.de, snitzer@redhat.com, zohar@linux.ibm.com
Subject: Re: [dm-devel] [PATCH 0/7] device mapper target measurements using IMA
Date: Mon,  2 Aug 2021 12:38:23 +0200	[thread overview]
Message-ID: <20210802103823.199091-1-public@thson.de> (raw)
In-Reply-To: <f73308f3-485f-46cb-0f20-6619edb541e6@linux.microsoft.com>

Hi Tushar,

thank you for answering my questions and looking at my suggestions.

> I can update the verity_status() to measure if v->signature_key_desc is 
> set.
> 
> Something like:
> DMEMIT("signature_key_desc_present=%c,", v->signature_key_desc ? 'y' :
> 'n');

If my understanding that this entry is only set if the signature was validated
is correct then this should work.

> Please note – even if we measure signature_key_desc (full string or just
> its presence): in order to use it with the keyrings, the IMA policy also
> needs to be set to measure key rings (using "measure func=KEY_CHECK
> ..."). It is independent from measuring the device mapper data (which is
> measured when the policy is set to “measure func=CRITICAL_DATA
> label=device-mapper ...").
> 
> Therefore measuring keyrings together (i.e. in the same IMA log) with DM
> data  is not always guaranteed, since it is dictated by how the IMA
> policy is configured.

Thanks for pointing that out. Currently we don't measure the keyrings but when
we enable remote attestation for dm-verity we'll make sure that our IMA policy
also measures the keyrings.

Regards,
Thore