From: Eric Wong <e@80x24.org>
To: Alyssa Ross <hi@alyssa.is>
Cc: meta@public-inbox.org
Subject: Re: Test failures in build sandbox
Date: Thu, 3 Oct 2019 01:28:07 +0000 [thread overview]
Message-ID: <20191003012807.GA52684@ailurophile> (raw)
In-Reply-To: <878sq2hd08.fsf@alyssa.is>
Alyssa Ross <hi@alyssa.is> wrote:
> Alyssa Ross <hi@alyssa.is> writes:
>
> > After some rather extensive debugging, I've determined that most of
> > these are to do with not being able to set the setgid bit. If I mask it
> > off in git's adjust_shared_perm function, I get down to two failures. I
> > don't know enough about filesystem permissions to know why this wouldn't
> > be allowed inside the sandbox if it is allowed normally, though.
>
> Aha! setuid and setgid permissions are disallowed in Nix builds using
> seccomp[1].
>
> So, I suppose it would be reasonable for us to either disable the test
> suite, or try to disable just the tests that try to create setgid
> things, unless you're open to adding a check for such a restriction and
> skipping tests if so. More than reasonable not to, though -- seccomp is
> a nightmare and next to impossible to reasonably support as upstream.
How do "git init --shared" tests work for the Nix git package? We could
try to follow that as an example...
>
> [1]: https://github.com/NixOS/nix/blob/5038e1bec43a71c97ae7f8be07218a8a2edbb6a1/src/libstore/build.cc#L2678
next prev parent reply other threads:[~2019-10-03 1:28 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-09-15 13:48 [PATCH] t/spawn: fix with unusual env locations Alyssa Ross
2019-09-15 18:55 ` Eric Wong
2019-09-24 1:11 ` Alyssa Ross
2019-09-24 4:01 ` Eric Wong
2019-09-24 22:12 ` Test failures in build sandbox Alyssa Ross
2019-09-26 8:44 ` Eric Wong
2019-10-02 23:23 ` Alyssa Ross
2019-10-03 0:46 ` Alyssa Ross
2019-10-03 1:28 ` Eric Wong [this message]
2019-10-03 1:43 ` Alyssa Ross
2019-10-03 7:57 ` [PATCH] t/search: bail out on `git init --shared' failures Eric Wong
2019-10-03 10:15 ` Alyssa Ross
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
List information: https://public-inbox.org/README
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20191003012807.GA52684@ailurophile \
--to=e@80x24.org \
--cc=hi@alyssa.is \
--cc=meta@public-inbox.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).