Re: [bug-inetutils] Re: [PATCH] Inetd and test scripts

[Mats Erik Andersson]

tisdag den  2 november 2010 klockan 23:09 skrev Ludovic Courtès detta:
> Hello,
> 
> Mats Erik Andersson <address@hidden> writes:
> 
> > --- a/ChangeLog
> > +++ b/ChangeLog
> > @@ -1,3 +1,11 @@
> > +2010-10-31  Mats Erik Andersson <address@hidden>
> > +
> > +   Improve the three test scripts.
> > +
> > +   * tests/tftp.sh: Make it robust and portable to GNU/Linux and BSD.
> > +   * tests/ping-localhost.sh: Test `ping' and `ping6'. Configurable 
> > targets.
> > +   * tests/traceroute-localhost.sh: Test UDP and ICMP. Configurable target.
> 
> [...]
> 
> > index 6b2dc80..f14f758 100755
> > --- a/tests/tftp.sh
> > +++ b/tests/tftp.sh
> > @@ -22,13 +22,15 @@
> >  TFTP="${TFTP:-../src/tftp$EXEEXT}"
> >  TFTPD="${TFTPD:-$PWD/../src/tftpd$EXEEXT}"
> >  INETD="${INETD:-../src/inetd$EXEEXT}"
> > -IFCONFIG="${IFCONFIG:-../ifconfig/ifconfig$EXEEXT}"
> > +IFCONFIG="${IFCONFIG:-../ifconfig/ifconfig$EXEEXT --format=unix}"
> >  
> > -PORT=7777
> > +AF=${AF:-inet}
> > +PROTO=${PROTO:-udp}
> > +PORT=${PORT:-7777}
> 
> Not sure these were absolutely necessary...

Only convenience for manual extension of test. I found it most useful.

> 
> [...]
> 
> > +# Check that the port is still available
> > +netstat -na | grep -q "^$PROTO .*$PORT "
> > +if test $? -eq 0; then
> > +    echo "Desired port $PORT/$PROTO is already in use."
> > +    exit 1
> > +fi
> 
> This breaks because ???netstat??? may not be in $PATH; in particular, it
> breaks the Hydra builder: <http://hydra.nixos.org/build/727555>.

True. Will be changed.

> 
> Furthermore, I think it???s useless: if ???inetd??? fails to listen on $PORT,
> the remainder of the script will notice it anyway.

Not at all. How do you then tell which part was failing: Inetd or Tftpd.

This is what robustness is all about. Predictable results, not ad hoc
failures.

> 
> Can you please fix it?
> 
> [...]
> 
> > +# Wait somewhat for the service to settle
> > +sleep 1
> 
> Pointless.

On the contrary, GNU/Linux was so quick to proceed that the test failed.
Again, robustness and predictable behaviour.

> > +# Did `inetd' really succeed in establishing a listener?
> > +netstat -na | grep "^$PROTO .*$PORT "
> > +if test $? -ne 0; then
> > +    # No it did not
> > +    ps "$inetd_pid" >/dev/null 2>&1 && kill "$inetd_pid"
> > +    echo "Failed in starting correct Inetd instance." >&2
> > +    exit 1
> > +fi
> 
> Same as above.

Same as above!!!!! This test is absolutely necessary. Otherwise
I could not have detected that Inetd failed to start in OpenBSD.

> 
> Thanks,
> Ludo???.
> 

Regards,
Mats