bug-automake
[Top][All Lists]
Advanced

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

Re: check4.test consistently failing


From: Ralf Wildenhues
Subject: Re: check4.test consistently failing
Date: Wed, 29 Apr 2009 21:01:50 +0200
User-agent: Mutt/1.5.18 (2008-05-17)

Hello Philipp,

* Philipp Thomas wrote on Wed, Apr 29, 2009 at 03:05:04PM CEST:
> * Ralf Wildenhues [20090428 22:42]:
> > Note that you can rerun this one test only with either
> >   cd tests
> >   make check VERBOSE=yes TESTS=check4.test

> I figured that out myself :) I'll do the test and will report. BTW, you can
> see the logs for yourself in the openSUSE buildservice. Just go to
> 
> https://build.opensuse.org/package/show?package=automake&project=openSUSE%3AFactory
> 
> and choose a log of any of the failing builds.

That "just" is an interesting statement.  I had to create an account
with Novell, then had to click through more pages.  I find the
buildservice quite complex at first, not knowing where the actual logs
would be.  Anyway, found at least one log with one failing test, but it
was check10.test not check4.test.

You need to search for
  ^FAIL:.*\.test$

to avoid being tricked by testsuite runs inside of testsuite runs.
Automake 1.11 logs will be easier to analyze (I hope).

The specific failure I found was due to /bin/sh being a ksh variant,
and fixed with this post-1.10.2 patch.

commit 319305db77dfdcbba3d5a1703c781750affdb20a
Author: Ralf Wildenhues <address@hidden>
Date:   Wed Nov 12 07:07:31 2008 +0100

    Testsuite fix for ksh.
    
    * tests/check10.test: Add ':' as last command in subshell, for
    zero exit status of the subshell.
    Missing backport from master reported by Tim Rice.
    


> > Second, do you have a testcase that shows what the require_file_internal
> > patch is required for?
> 
> See https://bugzilla.novell.com/show_bug.cgi?id=64822

Even with an account, I am not allowed to see this.  Wonder what kind of
policy is behind that.

> The problem AFAICS from the sources of patchutils 0.2.30 was that
> configure.in had

Can you please send a full bug report containing all necessary
information from above PR to bug-automake, including a reproducible test
case, in a separate thread?  Thanks.

Cheers,
Ralf




reply via email to

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