Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
From:   Steve Grubb <sgrubb@redhat.com>
To:     =?ISO-8859-1?Q?Wei=DF=2C?= Michael 
        <michael.weiss@aisec.fraunhofer.de>,
        Richard Guy Briggs <rgb@redhat.com>
Cc:     "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        "linux-audit@redhat.com" <linux-audit@redhat.com>,
        "casey@schaufler-ca.com" <casey@schaufler-ca.com>,
        "agk@redhat.com" <agk@redhat.com>,
        "linux-raid@vger.kernel.org" <linux-raid@vger.kernel.org>,
        "song@kernel.org" <song@kernel.org>,
        "eparis@redhat.com" <eparis@redhat.com>,
        "snitzer@redhat.com" <snitzer@redhat.com>,
        "paul@paul-moore.com" <paul@paul-moore.com>,
        "dm-devel@redhat.com" <dm-devel@redhat.com>
Subject: Re: [PATCH v4 0/3] dm: audit event logging
Date:   Wed, 08 Sep 2021 11:39:02 -0400
Message-ID: <4344604.LvFx2qVVIh@x2>
Organization: Red Hat
In-Reply-To: <20210908131616.GK490529@madcap2.tricolour.ca>
References: <20210904095934.5033-1-michael.weiss@aisec.fraunhofer.de> <9ca855cb19097b6fa98f2b3419864fd8ddadf065.camel@aisec.fraunhofer.de> <20210908131616.GK490529@madcap2.tricolour.ca>
MIME-Version: 1.0
Content-Transfer-Encoding: 7Bit
Content-Type: text/plain; charset="us-ascii"
Precedence: bulk

On Wednesday, September 8, 2021 9:16:16 AM EDT Richard Guy Briggs wrote:
>  Another minor oddity is the double "=" for the subj
> 
> > > field, which doesn't appear to be a bug in your code, but still
> > > puzzling.
> > 
> > In the test setup, I had Apparmor enabled and set as default security
> > module. This behavior occurs in any audit_log message.
> > Seems that this is coming from the label handling there. Having a quick
> > look at the code there is that they use '=' in the label to provide a
> > root view as part of their policy virtualization. The corresponding
> > commit is sitting there since 2017:
> > "26b7899510ae243e392960704ebdba52d05fbb13"
> 
> Interesting...  Thanks for tracking down that cause.  I don't know how
> much pain that will cause the userspace parsing tools.  I've added Steve
> Grubb to the Cc: to get his input, but this should not derail this patch
> set.

It likely breaks any parser. I would even say that it's a malformed event 
that should be corrected. There's been a published a specification for audit 
events  for at least 5 years. Latest copy is here:

https://github.com/linux-audit/audit-documentation/wiki/SPEC-Writing-Good-Events

-Steve