login
Search
Marcelo Tosatti has released the final 2.4.24 stable Linux kernel unchanged from 2.4.24-rc1. The main reason for the release was a local vulnerability in mremap() syscall, that allows local users to gain root privileges.
It's recommended that all users upgrade their kernel.
The upgraded 2.4.24 kernel release is available from numerous kernel.org mirrors.
For more information on this newly identified vulnerability, please read the announcement.
From: Marcelo Tosatti [email blocked] To: linux-kernel Subject: linux-2.4.24 released Date: Mon, 5 Jan 2004 05:55:57 -0800 final: - 2.4.24-rc1 was released as 2.4.24 with no changes. Summary of changes from v2.4.23 to v2.4.24-rc1 ============================================ <bjorn.helgaas:[blocked]>: o Fix 2.4 EFI RTC oops <marcelo.tosatti:[blocked]>: o Andrea Arcangeli: malicious users of mremap() syscall can gain priviledges <marcelo:[blocked]>: o Harald Welte: Fix ipchains MASQUERADE oops o Change EXTRAVERSION to 2.4.24-rc1 <trini:[blocked]>: o /dev/rtc can leak parts of kernel memory to unpriviledged users Jean Tourrilhes: o IrDA kernel log buster
xfs?
Shouldn't 2.4.24 contain xfs?
/me wondering
2.4.24 has xfs
It does. The changelog above is for 2.4.24-rc1 to 2.4.24 only.
it does not
look on the patch
no it has _not_
2.4.24 is a security release - the whole changelog from 2.4.23 to 2.4.24 is above. pre changes that were commited are postponed for 2.4.25-pre. So we'll see xfs and all in 2.4.25.
More details
http://isec.pl/vulnerabilities/isec-0013-mremap.txt has more details
2.2 and 2.6?
The link you gave says that also 2.2 and 2.6 are affected. Is there more information about this? 2.6 seems to have the same code so I think it is vulvenerable. How about 2.2? It has different code.
The bug is in mm/mremap.c, and this part of the 2.4.24 patch seems to fix it.
2.6 too
Yes, 2.6.0 is vulnerable. A fix just went in and we'll see it soon, probably 2.6.1rc2.
patch
The patch can be found here.
Linus said that -rc2 will have it.
On lkml discussion goes on how is 2.2 exploitable too, as it seems not...
2.2
It's prooved now 2.2 is not exploitable.
Openwall?
Out of curiosity, does anybody know if the Openwall kernel is also affected by this security vulnerability?
Re: Openwall?
Do you know a recent kernel root exploit that did not work on Openwall kernel?
this time marcelo has been faster
In his interview Marcelo said that the people of kerneltrap only blame him.
I want to say that this time I really appreciate this new and fast kernel relase to fix this bugs without having to wait too long.
Good!
2.4.24-ck1
For those that care, the 2.4.24-ck1 patchset is available.
http://www.plumlocosoft.com/kernel/