linux-kernel - %e in /proc/sys/kernel/core_pattern and empty current->comm

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Date:	Sun, 28 Aug 2011 09:45:08 -0700
From:	Earl Chew <echew@...acom.com>
To:	<linux-kernel@...r.kernel.org>
Subject: %e in /proc/sys/kernel/core_pattern and empty current->comm

It is possible to set /proc/sys/kernel/core_pattern to %e:

	echo '%e' > /proc/sys/kernel/core_pattern

A couple of interesting effects:

a. Cores will tend to try to overwrite the executable.

    Is this considered an issue ?

b. Apparently it seems possible to set current->comm to be empty:

	prctl(PR_SET_NAME, "", 0, 0, 0);

   No core dump is generated since the resulting corename is empty,
   and filp_open() fails.

   Is this considered a defect, or potential security concern ?

c. Consequential to (b), setting /proc/sys/kernel/core_pattern to empty
   and /proc/sys/kernel/core_uses_pid to zero universally prevents
   core dumps.

   I'm wondering if this behaviour is provided intentionally.

Earl

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/