Not OK: perl v5.7.0 +DEVEL7978 on armv4l-linux 2.2.17-rmk1 (UNINSTALLED)

[p5pRT]

Migrated from rt.perl.org#4853 (status was 'resolved')

Searchable as RT4853$

[p5pRT]

From @nwc10

lib/net-hostent......FAILED at test 6

/etc/hosts says
127.0.0.1 loopback localhost

known failure.

Is localhost in the aliases for gethostbyname ('localhost') on Win32?
If so, I think I can make a regression test that copes.

Nicholas Clark

Perl Info

Flags:
    category=install
    severity=none

Site configuration information for perl v5.7.0:

Configured by nick at Tue Dec  5 19:58:11 GMT 2000.

Summary of my perl5 (revision 5.0 version 7 subversion 0) configuration:
  Platform:
    osname=linux, osvers=2.2.17-rmk1, archname=armv4l-linux
    uname='linux bagpuss.unfortu.net 2.2.17-rmk1 #6 sat dec 2 19:44:40 gmt 2000 armv4l unknown '
    config_args='-Dusedevel -Ubincompat5005 -Uinstallusrbinperl -Dcf_email=nick@talking.bollo.cx -Dperladmin=nick@talking.bollo.cx -Dinc_version_list=  -Dinc_version_list_init=0 -Duseperlio -Dcc=/usr/local/bin/gcc -de'
    hint=recommended, useposix=true, d_sigaction=define
    usethreads=undef use5005threads=undef useithreads=undef usemultiplicity=undef
    useperlio=define d_sfio=undef uselargefiles=define usesocks=undef
    use64bitint=undef use64bitall=undef uselongdouble=undef
  Compiler:
    cc='/usr/local/bin/gcc', ccflags ='-fno-strict-aliasing -I/usr/local/include -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64',
    optimize='-O2',
    cppflags='-fno-strict-aliasing -I/usr/local/include'
    ccversion='', gccversion='2.95.2 20000516 (release) [Rebel.com]', gccosandvers=''
    intsize=4, longsize=4, ptrsize=4, doublesize=8, byteorder=1234
    d_longlong=define, longlongsize=8, d_longdbl=define, longdblsize=8
    ivtype='long', ivsize=4, nvtype='double', nvsize=8, Off_t='off_t', lseeksize=8
    alignbytes=4, usemymalloc=n, prototype=define
  Linker and Libraries:
    ld='/usr/local/bin/gcc', ldflags =' -L/usr/local/lib'
    libpth=/usr/local/lib /lib /usr/lib
    libs=-lnsl -lndbm -ldb -ldl -lm -lc -lposix -lcrypt -lutil
    perllibs=-lnsl -ldl -lm -lc -lposix -lcrypt -lutil
    libc=/lib/libc-2.1.3.so, so=so, useshrplib=false, libperl=libperl.a
  Dynamic Linking:
    dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun=undef, ccdlflags='-rdynamic'
    cccdlflags='-fpic', lddlflags='-shared -L/usr/local/lib'

Locally applied patches:
    DEVEL7978


@INC for perl v5.7.0:
    lib
    /usr/local/lib/perl5/5.7.0/armv4l-linux
    /usr/local/lib/perl5/5.7.0
    /usr/local/lib/perl5/site_perl/5.7.0/armv4l-linux
    /usr/local/lib/perl5/site_perl/5.7.0
    /usr/local/lib/perl5/site_perl
    .


Environment for perl v5.7.0:
    HOME=/home/nick
    LANG (unset)
    LANGUAGE (unset)
    LC_CTYPE=en_GB.ISO-8859-1
    LD_LIBRARY_PATH (unset)
    LOGDIR (unset)
    PATH=/home/nick/bin:/usr/local/bin:/usr/bin:/bin:/usr/bin/X11:/usr/games:/sbin:/usr/sbin:/usr/local/sbin
    PERL_BADLANG (unset)
    SHELL=/bin/sh

[p5pRT]

From [Unknown Contact. See original ticket]

Today around 10​:10pm, Nicholas Clark hammered out this masterpiece​:

:
: This is a build failure report for perl from nick@​talking.bollo.cx,
: generated with the help of perlbug 1.33 running under perl v5.7.0.
:
:
: -----------------------------------------------------------------
: [Please enter your report here]
:
: lib/net-hostent......FAILED at test 6
:
: /etc/hosts says
: 127.0.0.1 loopback localhost
:
: known failure.
:
: Is localhost in the aliases for gethostbyname ('localhost') on Win32?
: If so, I think I can make a regression test that copes.

Another slightly related problem I have is that 'localhost' isn't
mapped on a box I have. I'm not quite sure what to do about these
tests since they fail. There's no way of knowing of the module is
broken or if 'localhost' isn't mapped. I'm up for suggestions.

[p5pRT]

From [Unknown Contact. See original ticket]

Today around 6​:17pm, Casey R. Tweten hammered out this masterpiece​:

: Today around 10​:10pm, Nicholas Clark hammered out this masterpiece​:
:
: :
: : This is a build failure report for perl from nick@​talking.bollo.cx,
: : generated with the help of perlbug 1.33 running under perl v5.7.0.
: :
: :
: : -----------------------------------------------------------------
: : [Please enter your report here]
: :
: : lib/net-hostent......FAILED at test 6
: :
: : /etc/hosts says
: : 127.0.0.1 loopback localhost
: :
: : known failure.
: :
: : Is localhost in the aliases for gethostbyname ('localhost') on Win32?
: : If so, I think I can make a regression test that copes.
:
: Another slightly related problem I have is that 'localhost' isn't
: mapped on a box I have. I'm not quite sure what to do about these
: tests since they fail. There's no way of knowing of the module is
: broken or if 'localhost' isn't mapped. I'm up for suggestions.

Me​: "Be more helpful Casey!"

The box is True64.

[p5pRT]

From @nwc10

On Tue, Dec 05, 2000 at 06​:17​:32PM -0500, Casey R. Tweten wrote​:

Another slightly related problem I have is that 'localhost' isn't
mapped on a box I have. I'm not quite sure what to do about these
tests since they fail. There's no way of knowing of the module is
broken or if 'localhost' isn't mapped. I'm up for suggestions.

so you get
ok 1
not ok 2
not ok 3
not ok 4
not ok 5
not ok 6

because Net​::hostent loads successfully, but that's it?

Maybe we don't ok on the first gethost('localhost'), but print a comment
and skip all the rest of the tests?
Can we go for one of the root servers? Hmm. but that would fail on your
box when it's not plugged into any network.

Nicholas Clark

[p5pRT]

From [Unknown Contact. See original ticket]

Today around 11​:22pm, Nicholas Clark hammered out this masterpiece​:

: On Tue, Dec 05, 2000 at 06​:17​:32PM -0500, Casey R. Tweten wrote​:
: > Another slightly related problem I have is that 'localhost' isn't
: > mapped on a box I have. I'm not quite sure what to do about these
: > tests since they fail. There's no way of knowing of the module is
: > broken or if 'localhost' isn't mapped. I'm up for suggestions.
:
: so you get
: ok 1
: not ok 2
: not ok 3
: not ok 4
: not ok 5
: not ok 6
:
: because Net​::hostent loads successfully, but that's it?

Yes, that's what I used to get.

Something must have changed within the last few patches that changed
this cuz I just rsync'd ( last one was 2-3 days ago IIRC ) and ran
the process w/o a hitch.

The thing is, I had to make io_sock/io_multihomed look for 127.0.0.1
instead of localhost to work on that box and now net-hostent.t works
just great.

Oh well, I guess.

[p5pRT]

From [Unknown Contact. See original ticket]

Nicholas Clark <nick@​talking.bollo.cx> writes​:

This is a build failure report for perl from nick@​talking.bollo.cx,
generated with the help of perlbug 1.33 running under perl v5.7.0.

-----------------------------------------------------------------
[Please enter your report here]

lib/net-hostent......FAILED at test 6

/etc/hosts says
127.0.0.1 loopback localhost

known failure.

Is localhost in the aliases for gethostbyname ('localhost') on Win32?

Win32 is not a uniform platform in this area. As I recall NT3.*, Win95,
NT4, Win98, ... are potentially different.
Also if I recall correctly we at least used to use the fact that Win32
returns the system's name when passed 127.0.0.1 to mimic "hostname".

[p5pRT]

From @rspier

Win32 is excluded from these tests from that very reason - but it
seems that there may be too many variations on this - and the tests
for 127.0.0.1 <-> localhost might turn out to have been a bad idea.

"NI" == Nick Ing-Simmons <nik@​tiuk.ti.com> writes​:

NI> Nicholas Clark <nick@​talking.bollo.cx> writes​:

This is a build failure report for perl from nick@​talking.bollo.cx,
generated with the help of perlbug 1.33 running under perl v5.7.0.

-----------------------------------------------------------------
[Please enter your report here]

lib/net-hostent......FAILED at test 6

/etc/hosts says 127.0.0.1 loopback localhost

known failure.

Is localhost in the aliases for gethostbyname ('localhost') on
Win32?

NI> Win32 is not a uniform platform in this area. As I recall NT3.*,
NI> Win95, NT4, Win98, ... are potentially different. Also if I
NI> recall correctly we at least used to use the fact that Win32
NI> returns the system's name when passed 127.0.0.1 to mimic
NI> "hostname".

[p5pRT]

From @nwc10

On Wed, Dec 06, 2000 at 10​:37​:03AM -0500, Robert Spier wrote​:

Win32 is excluded from these tests from that very reason - but it
seems that there may be too many variations on this - and the tests
for 127.0.0.1 <-> localhost might turn out to have been a bad idea.

Is it safe to test to see that
1 lookup on "127.0.0.1" doesn't fail?
2 if lookup on "localhost" passes that it matches the result for the lookup
  on "127.0.0.1"?
2a or check that if "localhost" returns an IP address of 127.0.0.1 that the
  result of the result match "localhost"'s result?

(can windows do a lookup on "localhost" and return the IP of a non loopback
interface? in which case we'd need 2a)

Nicholas Clark

"NI" == Nick Ing-Simmons <nik@​tiuk.ti.com> writes​:

NI> Nicholas Clark <nick@​talking.bollo.cx> writes​:

This is a build failure report for perl from nick@​talking.bollo.cx,
generated with the help of perlbug 1.33 running under perl v5.7.0.

-----------------------------------------------------------------
[Please enter your report here]

lib/net-hostent......FAILED at test 6

/etc/hosts says 127.0.0.1 loopback localhost

known failure.

Is localhost in the aliases for gethostbyname ('localhost') on
Win32?

NI> Win32 is not a uniform platform in this area. As I recall NT3.*,
NI> Win95, NT4, Win98, ... are potentially different. Also if I
NI> recall correctly we at least used to use the fact that Win32
NI> returns the system's name when passed 127.0.0.1 to mimic
NI> "hostname".

[p5pRT]

From @rspier

"NC" == Nicholas Clark <nick@​ccl4.org> writes​:
NC> On Wed, Dec 06, 2000 at 10​:37​:03AM -0500, Robert Spier wrote​:

Win32 is excluded from these tests from that very reason - but it
seems that there may be too many variations on this - and the tests
for 127.0.0.1 <-> localhost might turn out to have been a bad idea.

NC> Is it safe to test to see that 1 lookup on "127.0.0.1" doesn't
NC> fail? 2 if lookup on "localhost" passes that it matches the
NC> result for the lookup on "127.0.0.1"? 2a or check that if
NC> "localhost" returns an IP address of 127.0.0.1 that the result of
NC> the result match "localhost"'s result?

NC> (can windows do a lookup on "localhost" and return the IP of a non
NC> loopback interface? in which case we'd need 2a)

I think it's safe to test for failure of a lookup - but for cases that
fall outside the /^localhost(?​:\..+)?$/ space, it could be almost
anything. loopback seems to be another case we might want to check.

I'm not so worried about 'passing' on Windows (as we do now), more
worried about the strange unices.

-R