Red Hat Bugzilla – Bug 1446288
CVE-2017-7482 kernel: net/rxrpc: overflow in decoding of krb5 principal
Last modified: 2018-08-28 18:16:46 EDT
When a kerberos 5 ticket is being decoded so that it can be loaded into an rxrpc-type key, the length of a variable-length field is checked to make sure that it's not going to overrun the allocated buffer space. The data is padded to the nearest four-byte boundary and the code doesn't check for this extra four-byte aligned padding. This can lead to the size-remaining variable wrapping and the data pointer accessing or reading past the end of the buffer. The read functionality could allow for a 3 byte infoleak and the write flaw could allow for an uncontrolled 3 byte write to kernels slab memory. This could lead to memory corruption and possible privilege escalation although no known exploit exists at the time of writing. Discussion an a CVE-ID assignment: http://seclists.org/oss-sec/2017/q2/602 Upstream fix https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=5f2f97656ada8d811d3c1bef503ced266fcd53a0
Acknowledgments: Name: Shi Lei (Qihoo 360 Gear Team)
Statement: This issue does not affect the Linux kernel packages as shipped with Red Hat Enterprise Linux 5,6 and 7 as the module is not enabled on these systems. This issue affects the Linux kernel packages as shipped with Red Hat Enterprise MRG-2. Future Linux kernel updates for the respective releases may address this issue.
Created kernel tracking bugs for this issue: Affects: fedora-all [bug 1464888]
As far as I can see in the kernel sources for the version 3.10.0-514.26.1 in RHEL 7, the affected code (from net/rxrpc/ar-key.c) is only built when CONFIG_AF_RXRPC is set. However, that kernel and at least some earlier ones have "# CONFIG_AF_RXRPC is not set" in their config files. So RHEL 7 is actually not affected by this issue, right? Or, perhaps, I am missing something.
You're 100% right, I had originally grepped the config for CONFIG_RX_RPC not CONFIG_AF_RX_RPC. Kernel-rt/MRG-2 still seems to use this as a module it remains affected. Closing trackers for 6 and 7 adjusting statement.
This was fixed for Fedora with 4.11.8 stable updates