To my surprise, not only are people using webmin, they still leave it open to the internet. [1]

[1] - https://www.shodan.io/search?query=webmin

Webmin probably has a higher than average number of users with limited or no technical skills, so not that surprising really.

This is why blaming the user isn't very effective, and safe defaults are so important.

Interesting that a search for Kubernetes on Shodan returns even more results than Webmin.

https://www.shodan.io/search?query=kubernetes

Webmin has had many remote vulnerabilities for many years, yet people still use it/leave it open to the internet.

https://www.cvedetails.com/vulnerability-list/vendor_id-358/...

> “It appears that a build/test system was compromised some time last year and the exploited added to code in the directory from which packages are built (and file timestamps modified to make this change not show up in a git diff),” Cameron said in an email.

How, exactly, does that happen?

Does git look at timestamps first and, assuming they aren't different, not hash the file contents to compare the SHA1? This [0] git FAQ makes me think the answer is "no".

Another FAQ:

> git diff and other Git operations is optimized so it does not even look at files whose status (size, modification time etc) on disk and in Git's index are different. [...] If the file has been touched somehow, git diff has to look at the content of and compare it which is a much slower operation even when there is in fact no change.

[0]: https://git.wiki.kernel.org/index.php/Git_FAQ#Why_isn.27t_Gi...

I suspect that in addition to modifying the timestamp, the attacker also had to ensure the filesizes were the same by possibly removing whitespaces or refactoring other code.

Is there a 'git verify' or something that will report such tampering?

Make sure to put any of your own personal web services behind some dumb authentication system to avoid most of these things biting you before you hear about them.

I use nginx + basic http auth, since I'm using it for https anyway.

Here's the original post from the author of Webmin: http://www.webmin.com/exploit.html

I hope they are able to find the source control logs and prosecute whoever embedded that code maliciously.

As mentioned (in large print!) in the article:

> “How the exploit happened is impossible to determine at this point, as the machine in question has been decommissioned."

I understand that, but they also seem to know a lot about what happened with respect to when things were checked in and with a little forensics perhaps they can figure out who it was. A deleted hard drive hasn't stopped people before and they shouldn't just give up that easily.

I feel like this is probably an inside job...

“and file timestamps modified to make this change not show up in a git diff”

...ok