Re: [gentoo-portage-dev] [PATCH 1/2] bin/install-qa-check.d: add new 90bad-bin-owner QA check.
On 07/29/2018 09:16 PM, Ulrich Mueller wrote: > > Staying with the man:man example, how would anybody become the "man" > user, in the first place? That user has /bin/false as a shell and no > valid password. One way would be to exploit a process that's running as the "man" user. Ostensibly such a thing is possible; otherwise we wouldn't be dropping privileges in the first place. The shell/password settings for that user are also in no way guaranteed. But on principle: who cares, non-root users shouldn't be able to gain root =) > Setgid executables shouldn't be group writable Why not? I'm happy to pull that part back out.
Re: [gentoo-portage-dev] [PATCH 1/2] bin/install-qa-check.d: add new 90bad-bin-owner QA check.
> On Sun, 29 Jul 2018, Michael Orlitzky wrote: > After thinking about this for a while, I think we should ignore setgid > but not setuid executables. The problem with setuid and a non-root owner > is that the owner can always exploit the situation: > Suppose /bin/foo is owned by "foo" and setuid. If root (or any other > privileged user) is about to run /bin/foo, then the "foo" user can > simply strip away the setuid bit and fill /bin/foo with malicious code. Staying with the man:man example, how would anybody become the "man" user, in the first place? That user has /bin/false as a shell and no valid password. > The same situation with setgid is safe because (as far as I know) > members of the group can't strip off the setgid bit. Setgid executables shouldn't be group writable, so I believe that part of the test is fine as-is in v1 of your patch. Ulrich pgppMBCz_RDB1.pgp Description: PGP signature
Re: [gentoo-portage-dev] [PATCH 1/2] bin/install-qa-check.d: add new 90bad-bin-owner QA check.
On 07/29/2018 03:43 PM, Ulrich Mueller wrote: > >> On a "normal" system, there is no good reason why the superuser should >> not own every system executable. This commit adds a new install-time >> check that reports any such binaries with a QA warning. To avoid false >> positives, non-"normal" systems (like prefix) are skipped at the moment. > > Shouldn't this check for setuid binaries like /usr/bin/mandb (which is > owned by man:man)? I think these are legitimate usage case. > After thinking about this for a while, I think we should ignore setgid but not setuid executables. The problem with setuid and a non-root owner is that the owner can always exploit the situation: Suppose /bin/foo is owned by "foo" and setuid. If root (or any other privileged user) is about to run /bin/foo, then the "foo" user can simply strip away the setuid bit and fill /bin/foo with malicious code. The same situation with setgid is safe because (as far as I know) members of the group can't strip off the setgid bit.
Re: [gentoo-portage-dev] [PATCH 1/2] bin/install-qa-check.d: add new 90bad-bin-owner QA check.
On 07/29/2018 03:43 PM, Ulrich Mueller wrote: > > Shouldn't this check for setuid binaries like /usr/bin/mandb (which is > owned by man:man)? I think these are legitimate usage case. > In general, yeah. I think we should be skeptical of setuid/gid executables, but this isn't the right place to make that stand. In this specific case, though, I don't see why that program is setuid. In fact, I'm pretty sure it lets the "man" user gain root privileges.
Re: [gentoo-portage-dev] [PATCH 1/2] bin/install-qa-check.d: add new 90bad-bin-owner QA check.
> On Sun, 29 Jul 2018, Michael Orlitzky wrote: > System executables that are not owned by root pose a security > risk. The owner of the executable is free to modify it at any time; > so, for example, he can change a daemon's behavior to make it > malicious before the next time the service is started (usually by > root). > On a "normal" system, there is no good reason why the superuser should > not own every system executable. This commit adds a new install-time > check that reports any such binaries with a QA warning. To avoid false > positives, non-"normal" systems (like prefix) are skipped at the moment. Shouldn't this check for setuid binaries like /usr/bin/mandb (which is owned by man:man)? I think these are legitimate usage case. Ulrich pgptqtrJo8E1U.pgp Description: PGP signature
Re: [gentoo-portage-dev] [PATCH 1/2] bin/install-qa-check.d: add new 90bad-bin-owner QA check.
W dniu nie, 29.07.2018 o godzinie 13∶37 -0400, użytkownik Michael
Orlitzky napisał:
> System executables that are not owned by root pose a security
> risk. The owner of the executable is free to modify it at any time;
> so, for example, he can change a daemon's behavior to make it
> malicious before the next time the service is started (usually by
> root).
>
> On a "normal" system, there is no good reason why the superuser should
> not own every system executable. This commit adds a new install-time
> check that reports any such binaries with a QA warning. To avoid false
> positives, non-"normal" systems (like prefix) are skipped at the moment.
>
> Bug: https://bugs.gentoo.org/629398
> ---
> bin/install-qa-check.d/90bad-bin-owner | 38
> ++
> 1 file changed, 38 insertions(+)
> create mode 100644 bin/install-qa-check.d/90bad-bin-owner
>
> diff --git a/bin/install-qa-check.d/90bad-bin-owner
> b/bin/install-qa-check.d/90bad-bin-owner
> new file mode 100644
> index 0..188d67a51
> --- /dev/null
> +++ b/bin/install-qa-check.d/90bad-bin-owner
> @@ -0,0 +1,38 @@
> +# Copyright 1999-2018 Gentoo Foundation
> +# Distributed under the terms of the GNU General Public License v2
> +
> +bad_bin_owner_check() {
> + # Warn about globally-installed executables (in /bin, /usr/bin, /sbin,
> + # or /usr/sbin) that are owned by a nonzero UID.
> +
> + # This check doesn't work on non-root prefix installations at
> + # the moment, because every executable therein is owned by a
> + # nonzero UID.
> + [[ "${EUID}" -ne "0" || "${PORTAGE_INST_UID}" -ne "0" ]] && return
> +
> + local d f found=()
> +
> + for d in "${ED%/}/bin" "${ED%/}/usr/bin" "${ED%/}/sbin"
> "${ED%/}/usr/sbin"; do
I think you should include /opt/bin as well. Or maybe simply all
locations on ${PATH}.
> + [[ -d "${d}" ]] || continue
> +
> + # Read the results of the "find" command into the "found" bash
> array.
> + # Use -L to catch symlinks whose targets are owned by a
> non-root user,
> + # even though it won't catch ABSOLUTE symlinks until the package
> + # is RE-installed (the first time around, the target won't
> exist).
> + while read -r -d '' f; do
> + found+=( "${f}" )
> + done < <(find -L "${d}" -maxdepth 1 -type f ! -uid 0 -print0)
> +
> + if [[ ${found[@]} ]]; then
> + eqawarn "system executables owned by nonzero uid:"
> + for f in "${found[@]}"; do
> + # Strip off the leading destdir before
> outputting the path,
> + # but leave the prefix if there is one.
> + eqawarn " ${f#${D%/}/}"
> + done
> + fi
> + done
> +}
> +
> +bad_bin_owner_check
> +:
--
Best regards,
Michał Górny
signature.asc
Description: This is a digitally signed message part
