bug-inetutils
[Top][All Lists]
Advanced

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

Re: inetutils 1.9.5 rc1 on Raspbian


From: Bruno Haible
Subject: Re: inetutils 1.9.5 rc1 on Raspbian
Date: Mon, 20 Jan 2020 01:46:04 +0100
User-agent: KMail/5.1.3 (Linux/4.4.0-171-generic; KDE/5.18.0; x86_64; ; )

Hi Mats,

> > But it does not indicate that tftp, syslogd, and telnet don't work.
> > Therefore I would find it better if the respective tests were marked
> > as "SKIP", not "FAIL".
> 
> Can you verify that macro TEST_IPV6 was set to `yes'?

TEST_IPV6 is set to 'auto', though config.status.

> It it not intended
> to be so, but I suspect that configuration did not manage to set the
> expected `TEST_IPV6 = no'. Should this have been the case, all three
> of tftp.sh, syslogd.sh, and telnet-localhost.sh are wrought to suppress
> every IPv6-addressing, thus skipping half of planned test cases.
> ... I have to figure out a configuration test
> to achieve this beyond simple header file poking.

On this platform (modern Linux, apparently just some system configuration
has IPv6 disabled) you cannot determine the absence of IPv6 at configure
time by looking at headers and library functions. It is a situation that
only becomes apparent while running the test.

What I do in this situation:
  1. Arrange the contents of the test in such a way that the mandatory
     stuff (IPv4) gets executed first; if it fails -> FAIL.
  2. Then comes the test of the optional features (IPv6). If it fails
     with a particular error indicating absence of IPv6 -> Either PASS
     or SKIP (as you like). If it fails in any other way -> FAIL.

Bruno




reply via email to

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