-
Notifications
You must be signed in to change notification settings - Fork 123
Add a security policy for Elektra #4214
Comments
Thank you very much for reporting! I agree that at least doc/AUTHORS.md should contain some information who may be contacted in which situations. |
@markus2330 As this was an issue for FLOSS last semester, we probably can also offer it for the current FLOSS course. |
No, I am afraid I need to do it at some point. It cannot be done by a FLOSS participant. |
Due do recent CVE-2022-3602 and CVE-2022-3786 i want to bring some attention to this topic again.
|
The topic is tagged as 0.9.* which means we definitely plan to do something before 1.0.0.
It describes how to do fuzz testing. What do you miss? I don't think it belongs to this issue, though.
Actually, afaik we already automatized all of it, so we can remove this sentence. @mpranj do you still do some manual tests? Can you remove the sentence?
I don't think it is very important:
But any reviews are very welcome. Which security audit were you thinking of? |
Basically I miss the transparency on what is exactly done during manual testing.
What about user software using Elektra? I don't know much about the internals but i could imagine it may be possible to somehow corrupt global mount db as non-admin user (e.g. override a mount for /etc/sudoers) or similar (overriding system/admin config from user namespace)
|
No, exactly this is impossible (at least as long the operating system works correctly).
Yes, but it wouldn't allow you to do admin actions, you get a permission error. See doc/TESTING.md for various workarounds to let the admin tests run on your computer. |
The recent log4j vulnerability (CVE-2021-44228) showed how important it is for a vulnerability to be be handled as quickly and as clean as possible.
This prompted me to search if Elektra has an established security policy in place but I couldn't find any.
If Elektra does not have such a policy, there are a few "standard" ways to create it, which would make it easier for researchers to report vulnerabilities:
SECURITY.md
file at the root of the project, which would describe how to do this.There is also a RFC draft which aims to create a standard on how this is done (tl;dr: a special security.txt file is placed under the
.well-known/
path on the website to define the security policy).It is important for vulnerabilities to be easily reportable as this going through a number of steps is seen as a chore and may discover researchers to report issues.
The text was updated successfully, but these errors were encountered: