ANNOUNCE: fping 3.16 (and version 4 pre-announcement)

[David Schweikert]

Hi,

I have just released fping 3.16 (see changelog below).

I also started working on a "fping 4.0" release, which will contain more
disruptive changes such as:

- unification of fping and fping6 binaries into one
- long options (in addition to the short options)

Considering that these changes might break some installations, I decided to
also maintain a "version 3" branch without these changes. Still, I hope that
the transition from fping version 3 to fping version 4 will be as smooth as
possible.

Changes since the last release:

2017-02-09 David Schweikert <da...@schweikert.ch>
* Version 3.16
* (feature) Support kernel-timestamping of received packets (#46)
* (feature) Simplify restrictions: only -i >= 1 and -p >= 10 are enforced now
* (bugfix) Fix option -m to return all IPs of a hostname
* (bugfix) Fix option -H (ttl) for IPv6
* (bugfix) Fix option -M (don't fragment) for IPv6
* (bugfix) Fix option -O (ToS) for IPv6
* (bugfix) Fix compatibility issue with AIX (#69, @blentzgh)
* (bugfix) Fix option -q not suppressing some ICMP error messages (#83)
* (bugfix) Fix option -M expecting an argument, when it shouldn't
* (bugfix) Fix minor issues found by Coverity Scan

You can download it here: http://fping.org/dist/

Cheers
David

[Axel Beckert]

Hi David,

David Schweikert schrieb am Thu, Feb 09, 2017 at 10:40:27AM +0100:
> I have just released fping 3.16 (see changelog below).

Thanks! I had uploaded 3.16-rc2 to Debian Experimental only yesterday.
Built fine on all release architectures. Only failed to build from
source on GNU/Hurd.

> I also started working on a "fping 4.0" release, which will contain more
> disruptive changes such as:
>
> - unification of fping and fping6 binaries into one

Ah, so despite that was in the 3.16-rc2, it's not in 3.16?

> Considering that these changes might break some installations, I decided to
> also maintain a "version 3" branch without these changes. Still, I hope that
> the transition from fping version 3 to fping version 4 will be as smooth as
> possible.

At least from the packaging side it was not much of an issue.

Gruss, Axel
--
/~\ Plain Text Ribbon Campaign | Axel Beckert
\ / Say No to HTML in E-Mail and News | a...@deuxchevaux.org (Mail)
X See http://arc.pasp.de/ | a...@noone.org (Mail+Jabber)
/ \ I love long mails: http://email.is-not-s.ms/ | http://abe.noone.org/ (Web)

[David Schweikert]

Hi Axel,

On Thu, Feb 09, 2017 at 12:30:17 +0100, Axel Beckert wrote:
> Thanks! I had uploaded 3.16-rc2 to Debian Experimental only yesterday.
> Built fine on all release architectures. Only failed to build from
> source on GNU/Hurd.

Cool, thanks. Do you know what the problem with GNU/Hurd is?

> > I also started working on a "fping 4.0" release, which will contain more
> > disruptive changes such as:
> >
> > - unification of fping and fping6 binaries into one
>
> Ah, so despite that was in the 3.16-rc2, it's not in 3.16?

Yes, correct. I thought that it would be better to release the
fping/fping6 unification changes as 4.0.

> > Considering that these changes might break some installations, I decided to
> > also maintain a "version 3" branch without these changes. Still, I hope that
> > the transition from fping version 3 to fping version 4 will be as smooth as
> > possible.
>
> At least from the packaging side it was not much of an issue.

Good to know, thanks.

Cheers
David

[Axel Beckert]

Hi David,

On Thu, Feb 09, 2017 at 12:39:25PM +0100, David Schweikert wrote:
> On Thu, Feb 09, 2017 at 12:30:17 +0100, Axel Beckert wrote:
> > Thanks! I had uploaded 3.16-rc2 to Debian Experimental only yesterday.
> > Built fine on all release architectures. Only failed to build from
> > source on GNU/Hurd.
>
> Cool, thanks. Do you know what the problem with GNU/Hurd is?

It seems to be this error:

gcc -DHAVE_CONFIG_H -I. -I.. -Wdate-time -D_FORTIFY_SOURCE=2 -Wall -Wextra -Wno-sign-compare -DIPV6 -g -O2 -fdebug-prefix-map=/<<PKGBUILDDIR>>=. -specs=/usr/share/dpkg/pie-compile.specs -fstack-protector-strong -Wformat -Werror=format-security -c -o fping-fping.o `test -f 'fping.c' || echo './'`fping.c
fping.c: In function 'main':
fping.c:612:60: error: 'IPV6_TCLASS' undeclared (first use in this function)
if ( setsockopt(socket6, IPPROTO_IPV6, IPV6_TCLASS, &tos, sizeof(tos))) {
^~~~~~~~~~~
fping.c:612:60: note: each undeclared identifier is reported only once for each function it appears in


Additionally there was this warning (not fatal, but IMHO noteworthy):

fping.c: In function 'receive_packet':
fping.c:1733:21: warning: unused variable 'cmsg' [-Wunused-variable]
struct cmsghdr *cmsg;
^~~~

Full log at
https://buildd.debian.org/status/fetch.php?pkg=fping&arch=hurd-i386&ver=3.16%7Erc2-1&stamp=1486588492&raw=0

Hints on how to solve common GNU/Hurd porting issues are at
https://www.gnu.org/software/hurd/hurd/porting/guidelines.html

(This specific issue seems not to be listed there, though, but it
might of help anyways.)

> > > I also started working on a "fping 4.0" release, which will contain more
> > > disruptive changes such as:
> > >
> > > - unification of fping and fping6 binaries into one
> >
> > Ah, so despite that was in the 3.16-rc2, it's not in 3.16?
>
> Yes, correct. I thought that it would be better to release the
> fping/fping6 unification changes as 4.0.

That's a good decision IMHO. A bit in the spirit of Semantic
Versioning (see http://semver.org/).

Kind regards, Axel

--
/~\ Plain Text Ribbon Campaign | Axel Beckert
\ / Say No to HTML in E-Mail and News | a...@deuxchevaux.org (Mail)

X See http://www.nonhtmlmail.org/campaign.html | a...@noone.org (Mail+Jabber)

[David Schweikert]

Hi Axel,

On Thu, Feb 09, 2017 at 12:57:02 +0100, Axel Beckert wrote:
> It seems to be this error:
>
> gcc -DHAVE_CONFIG_H -I. -I.. -Wdate-time -D_FORTIFY_SOURCE=2 -Wall -Wextra -Wno-sign-compare -DIPV6 -g -O2 -fdebug-prefix-map=/<<PKGBUILDDIR>>=. -specs=/usr/share/dpkg/pie-compile.specs -fstack-protector-strong -Wformat -Werror=format-security -c -o fping-fping.o `test -f 'fping.c' || echo './'`fping.c
> fping.c: In function 'main':
> fping.c:612:60: error: 'IPV6_TCLASS' undeclared (first use in this function)
> if ( setsockopt(socket6, IPPROTO_IPV6, IPV6_TCLASS, &tos, sizeof(tos))) {
> ^~~~~~~~~~~
> fping.c:612:60: note: each undeclared identifier is reported only once for each function it appears in

This should be fixed by this commit:
https://github.com/schweikert/fping/commit/b079eaf8d448585f2fc5e971e1207a9f5c85d4bf

> Additionally there was this warning (not fatal, but IMHO noteworthy):
>
> fping.c: In function 'receive_packet':
> fping.c:1733:21: warning: unused variable 'cmsg' [-Wunused-variable]
> struct cmsghdr *cmsg;
> ^~~~

Should be fixed by this commit:
https://github.com/schweikert/fping/commit/08e5cabe451b202800c1749f61a6bc62eaba6886

Thanks!

Cheers
David

[Axel Beckert]

Hi David,

I'm lagging a little bit with the Debian uploads. I just uploaded 3.16
to experimental (due to the freeze) despite 4.0-rc1 is out already.
Will upload that one (or another RC if released until then) in the
next few days, I hope.

I included both these fixes in the 3.16 upload and fping on Debian
GNU/Hurd builds now again:

https://buildd.debian.org/status/package.php?p=fping&suite=experimental