bind EADDRNOTAVAIL at wrong postition

bind EADDRNOTAVAIL at wrong postition

[Philipp Takacs]

Hi

In bind(2) there are general errors and errors specific for AF_UNIX. The
EADDRNOTAVAIL error is sorted at the AF_UNIX specific errors. But in
posix[0] EADDRNOTAVAIL is a normal error.

Philipp

[0] https://pubs.opengroup.org/onlinepubs/9699919799/functions/bind.html

Re: bind EADDRNOTAVAIL at wrong postition

[Alejandro Colomar]

[-- Attachment #1: Type: text/plain, Size: 635 bytes --]

Hi Philipp,

On Sun, Oct 06, 2024 at 03:17:48AM +0200, Philipp Takacs wrote:
> Hi
> 
> In bind(2) there are general errors and errors specific for AF_UNIX. The
> EADDRNOTAVAIL error is sorted at the AF_UNIX specific errors. But in
> posix[0] EADDRNOTAVAIL is a normal error.

Do you know any system under which it happens with a socket other than
AF_UNIX?  It would be interesting to know if that's just a POSIX thing,
or if actual systems do have that.

Have a lovely day!
Alex

> 
> Philipp
> 
> [0] https://pubs.opengroup.org/onlinepubs/9699919799/functions/bind.html

-- 
<https://www.alejandro-colomar.es/>

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: bind EADDRNOTAVAIL at wrong postition

[Philipp Takacs]

Hi Alex

[2024-11-01 13:48] Alejandro Colomar <alx@kernel.org>
> Hi Philipp,
>
> On Sun, Oct 06, 2024 at 03:17:48AM +0200, Philipp Takacs wrote:
> > Hi
> > 
> > In bind(2) there are general errors and errors specific for AF_UNIX. The
> > EADDRNOTAVAIL error is sorted at the AF_UNIX specific errors. But in
> > posix[0] EADDRNOTAVAIL is a normal error.
>
> Do you know any system under which it happens with a socket other than
> AF_UNIX?  It would be interesting to know if that's just a POSIX thing,
> or if actual systems do have that.

I found this while debugging a bug[0] on in the OpenSMTPD package of
Debian. The problem was that the interface was not fully up but already
had IP(v6) addresses configured. So OpenSMTPD get the addresses from
the Interface and the bind failed with EADDRNOTAVAIL.

There are also some other places where the linux kernel returns
EADDRNOTAVAIL on non AF_UNIX sockets, for example in sctp[1].

Philipp

[0] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1059700
[1] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/net/sctp/socket.c#n399

>
> Have a lovely day!
> Alex
>
> > 
> > Philipp
> > 
> > [0] https://pubs.opengroup.org/onlinepubs/9699919799/functions/bind.html
>
> -- 
> <https://www.alejandro-colomar.es/>
> part 2     application/pgp-signatur   833

Re: bind EADDRNOTAVAIL at wrong postition

[Alejandro Colomar]

[-- Attachment #1: Type: text/plain, Size: 1684 bytes --]

Hi Philipp.

On Tue, Nov 05, 2024 at 03:54:36PM GMT, Philipp Takacs wrote:
> Hi Alex
> 
> [2024-11-01 13:48] Alejandro Colomar <alx@kernel.org>
> > Hi Philipp,
> >
> > On Sun, Oct 06, 2024 at 03:17:48AM +0200, Philipp Takacs wrote:
> > > Hi
> > > 
> > > In bind(2) there are general errors and errors specific for AF_UNIX. The
> > > EADDRNOTAVAIL error is sorted at the AF_UNIX specific errors. But in
> > > posix[0] EADDRNOTAVAIL is a normal error.
> >
> > Do you know any system under which it happens with a socket other than
> > AF_UNIX?  It would be interesting to know if that's just a POSIX thing,
> > or if actual systems do have that.
> 
> I found this while debugging a bug[0] on in the OpenSMTPD package of
> Debian. The problem was that the interface was not fully up but already
> had IP(v6) addresses configured. So OpenSMTPD get the addresses from
> the Interface and the bind failed with EADDRNOTAVAIL.
> 
> There are also some other places where the linux kernel returns
> EADDRNOTAVAIL on non AF_UNIX sockets, for example in sctp[1].

Hmmm, then it looks like a documentation bug.  Would you mind sending a
patch?

Have a lovely night!
Alex

> 
> Philipp
> 
> [0] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1059700
> [1] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/net/sctp/socket.c#n399
> 
> >
> > Have a lovely day!
> > Alex
> >
> > > 
> > > Philipp
> > > 
> > > [0] https://pubs.opengroup.org/onlinepubs/9699919799/functions/bind.html
> >
> > -- 
> > <https://www.alejandro-colomar.es/>
> > part 2     application/pgp-signatur   833

-- 
<https://www.alejandro-colomar.es/>

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: bind EADDRNOTAVAIL at wrong postition

[Philipp Takacs]

Hi Alex,

[2024-11-05 23:34] Alejandro Colomar <alx@kernel.org>
> Hi Philipp.
>
> On Tue, Nov 05, 2024 at 03:54:36PM GMT, Philipp Takacs wrote:
> > Hi Alex
> > 
> > [2024-11-01 13:48] Alejandro Colomar <alx@kernel.org>
> > > Hi Philipp,
> > >
> > > On Sun, Oct 06, 2024 at 03:17:48AM +0200, Philipp Takacs wrote:
> > > > Hi
> > > > 
> > > > In bind(2) there are general errors and errors specific for AF_UNIX. The
> > > > EADDRNOTAVAIL error is sorted at the AF_UNIX specific errors. But in
> > > > posix[0] EADDRNOTAVAIL is a normal error.
> > >
> > > Do you know any system under which it happens with a socket other than
> > > AF_UNIX?  It would be interesting to know if that's just a POSIX thing,
> > > or if actual systems do have that.
> > 
> > I found this while debugging a bug[0] on in the OpenSMTPD package of
> > Debian. The problem was that the interface was not fully up but already
> > had IP(v6) addresses configured. So OpenSMTPD get the addresses from
> > the Interface and the bind failed with EADDRNOTAVAIL.
> > 
> > There are also some other places where the linux kernel returns
> > EADDRNOTAVAIL on non AF_UNIX sockets, for example in sctp[1].
>
> Hmmm, then it looks like a documentation bug.  Would you mind sending a
> patch?

Actualy I tried to avoid this, because I'm not a big fan of the
git-send-email workflow. But a patch is on the mailinglist.

Philipp

> Have a lovely night!
> Alex
>
> > 
> > Philipp
> > 
> > [0] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1059700
> > [1] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/net/sctp/socket.c#n399
> > 
> > >
> > > Have a lovely day!
> > > Alex
> > >
> > > > 
> > > > Philipp
> > > > 
> > > > [0] https://pubs.opengroup.org/onlinepubs/9699919799/functions/bind.html
> > >
> > > -- 
> > > <https://www.alejandro-colomar.es/>
> > > part 2     application/pgp-signatur   833
>
> -- 
> <https://www.alejandro-colomar.es/>

Re: bind EADDRNOTAVAIL at wrong postition

[Alejandro Colomar]

[-- Attachment #1: Type: text/plain, Size: 470 bytes --]

Hi Philip,

On Wed, Nov 06, 2024 at 11:55:58AM GMT, Philipp Takacs wrote:
> > Hmmm, then it looks like a documentation bug.  Would you mind sending a
> > patch?
> 
> Actualy I tried to avoid this, because I'm not a big fan of the
> git-send-email workflow.

Is the issue with sending patches as email, or specific to
git-send-email(1)?

> But a patch is on the mailinglist.

Thanks!

Have a lovely day!
Alex

-- 
<https://www.alejandro-colomar.es/>

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: bind EADDRNOTAVAIL at wrong postition

[Philipp Takacs]

[2024-11-06 12:02] Alejandro Colomar <alx@kernel.org>
> Hi Philip,
>
> On Wed, Nov 06, 2024 at 11:55:58AM GMT, Philipp Takacs wrote:
> > > Hmmm, then it looks like a documentation bug.  Would you mind sending a
> > > patch?
> > 
> > Actualy I tried to avoid this, because I'm not a big fan of the
> > git-send-email workflow.
>
> Is the issue with sending patches as email, or specific to
> git-send-email(1)?

It's git-send-email(1) specific. This has two reasons:

1. I don't think my vcs should include a mail user agent

2. I would prefere to just add commits as mime attachments

Philipp

> > But a patch is on the mailinglist.
>
> Thanks!
>
> Have a lovely day!
> Alex
>
> -- 
> <https://www.alejandro-colomar.es/>

Re: bind EADDRNOTAVAIL at wrong postition

[Alejandro Colomar]

[-- Attachment #1: Type: text/plain, Size: 1416 bytes --]

Hi Philipp,

On Wed, Nov 06, 2024 at 08:47:04PM GMT, Philipp Takacs wrote:
> > > Actualy I tried to avoid this, because I'm not a big fan of the
> > > git-send-email workflow.
> >
> > Is the issue with sending patches as email, or specific to
> > git-send-email(1)?
> 
> It's git-send-email(1) specific. This has two reasons:
> 
> 1. I don't think my vcs should include a mail user agent

You don't need it.  In fact, I use neomutt(1) as a driver for
git-send-email(1), which leaves the MUA job to neomutt(1).
git-send-email(1) itself doesn't do much work.

Here's what I use:

$ sed -n '/^\[sendemail]/,/^[^\s]/p' ~/.gitconfig 
[sendemail]
	sendmailcmd = neomutt -C -H - && true

In fact, I prefer if a real MUA does the send, because then it allows
signing the emails with PGP.

> 2. I would prefere to just add commits as mime attachments

I accept them as mime attachments.  But I strongly prefer having one
patch per email.  If it's okay for you to send one email per patch, and
want to do it manually, I don't care too much if it's a mime attachment
or inline.  git-send-email(1) is nice in that regard, because it
automates invoking the MUA and piping one patch to each invocation.

You can (and probably have already done) check the files under
./CONTRIBUTING.d/, which detail this.

> Philipp

Have a lovely night!
Alex


-- 
<https://www.alejandro-colomar.es/>

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]