Bugzilla – Bug 389152
VUL-0: CVE-2008-2136: kernel: remote dos fixed in 2.6.25.3
Last modified: 2018-10-03 07:52:41 UTC
Your friendly security team received the following report. Please respond ASAP. The issue is public. Date: Sun, 11 May 2008 12:32:08 +0200 From: Andi Kleen <andi@firstfloor.org> To: security-intern@suse.de Subject: [sec-int] remote dos fixed in 2.6.25.3 Just in case you missed it the kfree_skb fix in 2.6.25.3 fixes a DoS that allows everyone to fill all memory as soon as the ipv6 module is loaded and SIT is not firewalled. Should probably get kernel security updates at high priority. -Andi Marcus says: gitlinks: http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.25.y.git;a=commit;h=36ca34cc3b8335eb1fe8bd9a1d0a2592980c3f02 http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.25.y.git;a=commitdiff;h=1be05a5eda841014c1151cb0f8dc791862bd40a5
code looks same down to SLES 9 .. so I guess all 2.6 are affected. Also SLES8 _rcv function looks the same, so I guess also 2.4. This is a SLES 8 patch candidate. ccing reporter ;)
Karsten, can you do this for all kernel versions?
There is a CVE requested for this, but it is not assigned yet, should happen tomorrow.
CVE-2008-2136
Greg, should I fix this for head too, or will we go to 2.6.25.3 soon ?
Did submit to kernel CVS: SLES8 SLES9 SP3,SP4 SLES10 SP1,SP2 (SP2 disabled with +post-sp2) SL102 SL103
(In reply to comment #6 from Karsten Keil) > Greg, should I fix this for head too, or will we go to 2.6.25.3 soon ? > We have already moved to 2.6.25.3 in HEAD, so it is fixed there already.
Argh, I didn't saw that cvs up did fail (temporary problem on my side), sorry for the noise.
So what's the state of this now? I cannot really believe that a ping of death type problem takes that long to fix. Would the release process go faster if I posted a exploit publicly?
no, it would not go faster, it would just increase my stress levels. But feel free to do anyway. :) A kernel update is in the making already, just no one seems to want to hurry.
Karsten, I do not see it in the SL103 branch. DId you really commit it there? it is in SL102, SLES10SP1, SLES10SP2 (this one is already released), SLES9SP4 and SLES8. => Only in SL103 missing.
Done, seems the local cvs issue from comment #9 also did not commit the changes to the main rep., they already were in my tree, sorry for the delay.
all done for kernel -> fixed
10.2 and 10.3 kernel updates released
released sles9 updates kernel version is: 2.6.5-7.312
This bug was mentioned / fixed in the SLERT 10 SP2 kernel, version 2.6.22.19-0.15, released Wed 23 July 2008.