Options to audit2allow and manpage phrasing

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Folks, I've been studying audit2allow in order to understand how to
use the sepolgen library, and I've a couple of questions about how the
code is structured and how the manpage is phrased. One concerns
dontaudit (-D), the other concerns refpolicy (-R/-N). I'd appreciate
your reviewing this and setting me straight.

For dontaudit, audit2allow reads

       parser.add_option("-D", "--dontaudit", action="store_true",
                          dest="dontaudit", default=False,
                          help="generate policy with dontaudit rules")

while  audit2allow.1 reads

    .B "\-D" | "\-\-dontaudit"
    Generate dontaudit rules (Default: allow)

Since '-D' defaults to false and only sets dontaudit true if present,
shouldn't the man page read

    .B "\-D" | "\-\-dontaudit"
    Generate dontaudit rules (Default: False, do not generate dontaudit rules)

???

They may mean the same thing, but the second reading seems clearer to me.

It's a bit muddier for refpolicy-style-output, for which audit2allow reads

        parser.add_option("-R", "--reference", action="store_true",
dest="refpolicy",
                          default=True, help="generate refpolicy style output")
        parser.add_option("-N", "--noreference", action="store_false",
dest="refpolicy",
                          default=False, help="do not generate
refpolicy style output")

The corresponding lines in audit2allow.1 are:

    .B "\-N" | "\-\-noreference"
    Do not generate reference policy, traditional style allow rules.
    This is the default behavior.

    .B "\-R" | "\-\-reference"
    Generate reference policy using installed macros.
    This attempts to match denials against interfaces and may be inaccurate.

Since both -R and -N set refpolicy, the only way I can make sense of
this is if defining the '-N' option *after* the definition of '-R'
overrides refpolicy as set by -R. That seems reasonable to me, but I
wanted to confirm. It might make it clearer if the -R option had a
default of false, as in

        parser.add_option("-R", "--reference", action="store_true",
dest="refpolicy",
                          default=False, help="generate refpolicy style output")

In other words, leave -R out, and default to the -N behaviour.

Since -R and -N are opposites, would it make sense to make them
mutually exclusive?

        refPolOpt = parser.add_mutually_exclusive_group()
        refPolOpt.add_option("-R", "--reference", action="store_true",
dest="refpolicy",
                          default=False, help="generate refpolicy style output")
        refPolOpt.add_option("-N", "--noreference",
action="store_false", dest="refpolicy",
                          default=False, help="do not generate
refpolicy style output")

This would make it clearer that only one of these options should be
provided and that only -R changes default behaviour.

Thoughts?

Thanks,

P

Peter Whittaker
EdgeKeep Inc.
www.edgekeep.com
+1 613 864 5337
+1 613 864 KEEP



[Index of Archives]     [Selinux Refpolicy]     [Linux SGX]     [Fedora Users]     [Fedora Desktop]     [Yosemite Photos]     [Yosemite Camping]     [Yosemite Campsites]     [KDE Users]     [Gnome Users]

  Powered by Linux