bug-guix
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

bug#27795: Issues with upstream source for guile-emacs


From: Christopher Allan Webber
Subject: bug#27795: Issues with upstream source for guile-emacs
Date: Sat, 29 Jul 2017 11:20:19 -0500
User-agent: mu4e 0.9.18; emacs 25.2.1

Leo Famulari writes:

> On Sun, Jul 23, 2017 at 04:22:06PM +0200, Ricardo Wurmus wrote:
>>
>> Ricardo Wurmus <address@hidden> writes:
>>
>> > Leo Famulari <address@hidden> writes:
>> >
>> >> While working on the bug 'Changing package source URLs from git:// to
>> >> https://' [0], I noticed an issue with the sources for guile-emacs.
>> >>
>> >> We currently fetch this source code over the unauthenticated GIT
>> >> protocol. It is also available over HTTPS. However, these two protocols
>> >> are returning different Git repos for some reason.
>> >
>> > The clone times out for me:
>> >
>> > --8<---------------cut here---------------start------------->8---
>> > git clone https://git.hcoop.net/git/bpt/emacs.git guile-emacs-over-https
>> > Cloning into 'guile-emacs-over-https'...
>> > ^C
>> > --8<---------------cut here---------------end--------------->8---
>> >
>> > But the clone from git:// works fine.
>> >
>> > Is the repository actually served over HTTPS?
>>
>> Don’t mind me.  It eventually worked.  The repositories have different
>> histories, and the https-repo looks like it is two commits behind.
>> Looks like an older rebase.
>>
>> I’d say we should leave it with the current git:// URL.
>
> The thing is, since the git:// protocol is unauthenticated, we could
> assume that those extra two commits are added by a MitM :/
>
> Somebody who is interested in guile-emacs should really ask upstream
> what is going on.

Since we hash the checkout's contents, an attacker would have to be very
consistently adding those two commits for both the original packager
(me) and all subsequent users... a possible attack, but I think it's not
the biggest thing to worry about.





reply via email to

[Prev in Thread] Current Thread [Next in Thread]