Always leave a trailing zero at the end of sun_path

[kalvdans]

According to the unix(7) manpage on Linux:

   When binding a socket to a pathname, a few rules should be observed for maximum portability and ease of coding:
   *  The pathname in sun_path should be null-terminated.
   *  The length of the pathname, including the terminating null byte, should not exceed the size of sun_path.
   *  The addrlen argument that describes the enclosing sockaddr_un structure should have a value of at least:
          offsetof(struct sockaddr_un, sun_path)+strlen(addr.sun_path)+1
      or, more simply, addrlen can be specified as sizeof(struct sockaddr_un).

[kalvdans]

I need some help to reproduce the CI failures. I could not reproduce on a mac at work:

insight@mac openssh-portable % mkdir -p /tmp/usr/local/etc/
insight@mac openssh-portable % ./ssh-keygen -A -f /tmp     
ssh-keygen: generating new host keys: ED25519 
insight@mac openssh-portable % sw_vers -productVersion                 
13.4

But then, the CI environment is quite different. Can you try on your end, @oliverkurth ?

[oliverkurth]

But then, the CI environment is quite different. Can you try on your end, @oliverkurth

I saw that, and tried to reproduce on my Mac, but it works fine for me too.

The CI is a mess, and I have failures for the unix socket PR that I am not able to reproduce. It's very frustrating.

[kalvdans]

But then, the CI environment is quite different. Can you try on your end, @oliverkurth

I saw that, and tried to reproduce on my Mac, but it works fine for me too.

Thanks a lot for trying to reproduce!

The CI is a mess, and I have failures for the unix socket PR that I am not able to reproduce. It's very frustrating.

I force-pushed to re-trigger CI, but the failure is consistent. Good that it isn't intermittent at least :)

[oliverkurth]

Looks like the segfault in ssh-keygen on MacOS is a known issue: https://bugzilla.mindrot.org/show_bug.cgi?id=3584

It's not your fault 🙂