bug-coreutils
[Top][All Lists]
Advanced

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

Re: coreutils-6.9.90 released


From: Jim Meyering
Subject: Re: coreutils-6.9.90 released
Date: Sun, 02 Dec 2007 16:29:14 +0100

Greg Schafer <address@hidden> wrote:
> Jim Meyering wrote:
>
>> Coreutils version 6.9.90 has been released.
>
> Thanks for your efforts.
>
>> Since the lzma-compressed tarball is 2GB smaller than the .bz2 one,
>> and since it decompresses twice as fast, I no longer create bzip2-
>> compressed tarballs.
>
> Ugh. Fair enough on promoting lzma, but -1 on dropping bzip2 altogether.

Why?  Is your bandwidth so limited that 8.6MB gzip-compressed tarball
is a problem?  I considered making only bzip2- and lzma-compressed
tarballs, but am pretty sure gzip still has the ubiquity advantage.

To reiterate, here are the sizes:
  coreutils-6.9.90.tar.gz    8.6 MB
  coreutils-6.9.90.tar.bz2   5.6 MB
  coreutils-6.9.90.tar.lzma  3.6 MB

If enough people miss bzip2 and there's a good case for it,
I'll resume making them.

>> ** New build options
>>
>>   By default, "make install" no longer attempts to install (or even build) 
>> su.
>>   To change that, use ./configure --enable-install-program=su.
>>   If you also want to install the new "arch" program, do this:
>>   ./configure --enable-install-program=arch,su.
>
> I actually prefer `su' from Coreutils so I'm glad it can still be
> installed. However, I notice a difference from 6.9 where installation of
> `su' was skipped when there was insufficient access. Now it just gets
> installed anyway (with a missing setuid bit).

Thanks for reporting that.
Patch below.

>>   You can inhibit the compilation and installation of selected programs
>>   at configure time.  For example, to avoid installing "hostname" and
>>   "uptime", use ./configure --enable-no-install-program=hostname,uptime
>>   Note: currently, "make check" passes, even when arch and su are not
>>   built (that's the new default).  However, if you inhibit the building
>>   and installation of other programs, don't be surprised if some parts
>>   of "make check" fail.
>
> Why break the test suite intentionally?

Priorities.
I have limited time for this work.
Making the test suite work for less-common (this is unusual, imho)
use cases is not as important as say, making sure "real" bugs
are fixed in a timely manner.

> This will likely bite a lot of
> folks and become a FAQ. I prefer to omit `kill' and `uptime' so now `make
> check' chokes thusly:
>
>         done;                                           \
>         exit $status
> 36a37
>> kill
> 91a93
>> uptime
> make[2]: *** [check-x-vs-1] Error 1

Thanks for reporting it.
Someone will get to it eventually.
If you send in a patch, it may be fixed in a release even sooner.

By the way, why do you prefer to omit those?

> Forcing everyone to use `make -k check' is not ideal IMHO. It would be
> nice if `--enable-no-install-program' could be made to play nicely with
> the test suite.
>
> BTW, I'm seeing 1 failure here in the gnulib testsuite:
>
> Finding www.gnu.org service http...
> res -3: Temporary failure in name resolution
> Finding www.ibm.com service https...
> res -3: Temporary failure in name resolution
> Finding microsoft.com service http...
> res -3: Temporary failure in name resolution
> Finding google.org service ldap...
> res -3: Temporary failure in name resolution
> FAIL: test-getaddrinfo
>
> It's clearly due to a missing `/etc/resolv.conf'. This is in the context
> of building up a new system inside a chroot environment where not
> everything is set up yet. It would be nice if this test could be skipped
> when there is no network.

I agree.
----------------
Here's the patch mentioned above.

        Remove just-installed su, if unable to make it set-UID root.

        Normally, su isn't even installed.  However, if you configure with
        --enable-install-program=su, and then install with insufficient
        privileges, we now make sure to remove the just-installed binary.
        * src/Makefile.am (install-exec-hook): Rename from install-exec-local,
        so that this rule is guaranteed to be run *after* installation.
        If unable to chown and chmod the installed "su" program, remove it.
        Reported by Greg Schaefer.

diff --git a/src/Makefile.am b/src/Makefile.am
index 49979fd..2f40a04 100644
--- a/src/Makefile.am
+++ b/src/Makefile.am
@@ -178,7 +178,7 @@ install_su =                                                
\
 install-root: su$(EXEEXT)
        @$(install_su)

-install-exec-local: su$(EXEEXT)
+install-exec-hook: su$(EXEEXT)
        @if test "$(INSTALL_SU)" = yes; then                    \
          TMPFILE=$(DESTDIR)$(bindir)/.su-$$$$;                 \
          rm -f $$TMPFILE;                                      \
@@ -195,6 +195,7 @@ install-exec-local: su$(EXEEXT)
          else                                                  \
            echo "WARNING: insufficient access; not installing su"; \
            echo "NOTE: to install su, run 'make install-root' as root"; \
+           rm -f $(installed_su);                              \
          fi;                                                   \
        else :;                                                 \
        fi
--
1.5.3.6.985.g65c6a4




reply via email to

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