[GH-ISSUE #869] Can you use dirtycow to break out of firejail? [ QUESTION] #591

Closed
opened 2026-05-05 06:14:14 -06:00 by gitea-mirror · 6 comments
Owner

Originally created by @suedi on GitHub (Oct 23, 2016).
Original GitHub issue: https://github.com/netblue30/firejail/issues/869

Can you use dirtycow to break out of firejail?

DirtyCow as in https://github.com/dirtycow/dirtycow.github.io/wiki/VulnerabilityDetails

Originally created by @suedi on GitHub (Oct 23, 2016). Original GitHub issue: https://github.com/netblue30/firejail/issues/869 Can you use dirtycow to break out of firejail? DirtyCow as in https://github.com/dirtycow/dirtycow.github.io/wiki/VulnerabilityDetails
gitea-mirror 2026-05-05 06:14:14 -06:00
Author
Owner

@Fred-Barclay commented on GitHub (Oct 23, 2016):

I'm working on testing that in a VM. 😄

<!-- gh-comment-id:255609043 --> @Fred-Barclay commented on GitHub (Oct 23, 2016): I'm working on testing that in a VM. :smile:
Author
Owner

@reinerh commented on GitHub (Oct 23, 2016):

The ptrace-based exploits are prevented with "firejail --seccomp", as this drops ptracing by default.
When --apparmor is used with the default apparmor firejail profile, there should also no access to /proc/self/mem, so those would also not work.

You could also try adding madvise to the list of syscalls dropped with seccomp for additional protection, though I'm not sure how often this is used legitimately.

<!-- gh-comment-id:255620710 --> @reinerh commented on GitHub (Oct 23, 2016): The ptrace-based exploits are prevented with "firejail --seccomp", as this drops ptracing by default. When --apparmor is used with the default apparmor firejail profile, there should also no access to /proc/self/mem, so those would also not work. You could also try adding madvise to the list of syscalls dropped with seccomp for additional protection, though I'm not sure how often this is used legitimately.
Author
Owner

@netblue30 commented on GitHub (Oct 25, 2016):

I would say you need to update the kernel.

<!-- gh-comment-id:256035033 --> @netblue30 commented on GitHub (Oct 25, 2016): I would say you need to update the kernel.
Author
Owner

@suedi commented on GitHub (Oct 26, 2016):

Yeah, that's wot I wos fraid of now I have to find time to re-compile the kernel :(

closing...

<!-- gh-comment-id:256454600 --> @suedi commented on GitHub (Oct 26, 2016): Yeah, that's wot I wos fraid of now I have to find time to re-compile the kernel :( closing...
Author
Owner

@Fred-Barclay commented on GitHub (Oct 26, 2016):

@suedi What distro are you using?

I know that Debian and Ubuntu (and Mint) have already got the fixes in their repos, so no need to recompile if you use either. I suspect most other distros that use installable binaries have updates too, so unless you use a distro that requires you to recompile you shouldn't have to.

<!-- gh-comment-id:256455143 --> @Fred-Barclay commented on GitHub (Oct 26, 2016): @suedi What distro are you using? I know that Debian and Ubuntu (and Mint) have already got the fixes in their repos, so no need to recompile if you use either. I suspect most other distros that use installable binaries have updates too, so unless you use a distro that requires you to recompile you shouldn't have to.
Author
Owner

@suedi commented on GitHub (Oct 27, 2016):

Yeah, on my rpi I already updated but on my other comps
I roll my own with AUFS patches so will need to re-compile me thinks

<!-- gh-comment-id:256714633 --> @suedi commented on GitHub (Oct 27, 2016): Yeah, on my rpi I already updated but on my other comps I roll my own with AUFS patches so will need to re-compile me thinks
Sign in to join this conversation.
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference: github-starred/firejail#591
No description provided.