The SNMP NAT helper netfilter can cause a memory corruption. The corruption can be triggered remotely when the ip_nat_snmp_basic module is loaded and traffic on port 161 or 162 is NATed. Proposed upstream fix can be found here: https://lists.netfilter.org/pipermail/netfilter-devel/2006-May/024422.html
Reapplying lost comment from jburke on 2006-06-12 15:14 EST: Building a test kernel now. Once built I will run the kernel through RHTS.
The RHTS job has finished but with errors. The errors are not because of the kernel. There were power outages in the RDU office that caused the tests to fail. I will reschedule the RHTS jobs for S390 and ppc. All other arches have passed.
I am still trying to get the original issue reproduced. It looks like it can be reproduced with a test tool from MIT: ftp://ftp.lcs.mit.edu/nets/snmpdk.tar.Z http://www.cis.upenn.edu/~davin/snmpdk.html I will first verify that I can duplicate the issue. If in fact I can verify the issue. I will then verify the fix by using the kernel that is being built through beehive kernel-2.4.21-44.EL.BZ192633.2.src.rpm.
java-based version of the snmp test tool available here: http://www.ee.oulu.fi/research/ouspg/protos/testing/c06/snmpv1
Downgrading impact to moderate due to "unlikely configuration".
Created attachment 131213 [details] proposed patch
A fix for this problem has just been committed to the RHEL3 U8 patch pool this evening (in kernel version 2.4.21-45.EL).
An advisory has been issued which should help the problem described in this bug report. This report is therefore being closed with a resolution of ERRATA. For more information on the solution and/or where to find the updated files, please follow the link below. You may reopen this bug report if the solution does not work for you. http://rhn.redhat.com/errata/RHSA-2006-0437.html