--- Begin Message ---
|
Subject: |
make check problems with coreutils 8.16 and earlier |
|
Date: |
Mon, 16 Apr 2012 12:48:13 -0500 (CDT) |
|
User-agent: |
Alpine 2.02 (SOC 1266 2009-07-14) |
[I'm not subscribed to bug-coretuils, please Cc: me on any relevant
replies]
Hi All!
I originally reported this issue with GNU coreutils 7.2 in 2009, start
of the thread is available here
http://lists.gnu.org/archive/html/bug-coreutils/2009-04/msg00182.html
The thread died out without any resolution, unfortunately.
The same issue has happened with every version of coreutils I've built
since then, including 8.16 that I just tried recently.
I'm using bash (currently 4.2.10) as my login shell. As mentioned in
the original thread referenced above, what appears to be happening is
that bash is being used to run all of the tests, and when my environment
is loaded the PATH from my environment is causing at least some of the
tests to get the OS versions of commands, rather than the versions from
the src directory that should be getting tested. This causes some painful
hangs and other general failures.
I have confirmed that if I do
gmake check VERBOSE=yes SHELL=/usr/xpg4/bin/sh
then the tests perform as expected -- I get four failures and one
error and quite a number of skips, but no issues with tests hanging
because it's the OS version of the command being run:
# TOTAL: 477
# PASS: 393
# SKIP: 79
# XFAIL: 0
# FAIL: 4
# XPASS: 0
# ERROR: 1
So, the problem appears to be that the test cases aren't sufficiently
insulated from the user's environment.
I build my PATH from scratch in my .bashrc, rather than inheriting it from
the system defaults. I'm guessing this is what's causing the test
failures.
Since I have the work-around of forcing a different SHELL for the tests
I can make do, but I thought I should at least report the issue as still
present. I would be willing to pursue it further, if anyone thinks it
should be pursued.
Tim
--
Tim Mooney address@hidden
Enterprise Computing & Infrastructure 701-231-1076 (Voice)
Room 242-J6, IACC Building 701-231-8541 (Fax)
North Dakota State University, Fargo, ND 58105-5164
--- End Message ---
--- Begin Message ---
|
Subject: |
Re: bug#11256: make check problems with coreutils 8.16 and earlier |
|
Date: |
Tue, 08 May 2012 11:04:26 +0200 |
Jim Meyering wrote:
> Tim Mooney wrote:
...
> I think your suggestion below will solve your remaining
> "make check" failures.
>
>> If you have any interest in it, I would consider supply a patch against
>> the tests that checks to see if the invoking shell is bash, and if it is,
>> it just runs "unalias -a" at the start of the test.
>
> Thanks for the suggestion.
> The file that needed the change comes from gnulib
> and is about to be synced from there to coreutils.
...
> Subject: [PATCH] init.sh: don't let bash aliases interfere with tests
With that, I'm closing this ticket.
--- End Message ---