automake-1.15-6.fc25 fails to build in F25: make check-DEJAGNU check-TESTS check-local make[1]: Entering directory '/builddir/build/BUILD/automake-1.15/t/check12.dir/build' Making a new site.exp file ... srcdir='..'; export srcdir; \ EXPECT=expect; export EXPECT; \ if /bin/sh -c "runtest --version" > /dev/null 2>&1; then \ exit_status=0; l='hammer spanner'; for tool in $l; do \ if runtest HAMMER=../hammer SPANNER=../spanner --tool $tool --srcdir $srcdir ; \ then :; else exit_status=1; fi; \ done; \ else echo "WARNING: could not find 'runtest'" 1>&2; :;\ fi; \ exit $exit_status WARNING: Couldn't find the global config file. WARNING: Couldn't find tool init file Test run by mockbuild on Wed Jun 22 14:49:05 2016 Native configuration is i686-pc-linux-gnu === hammer tests === Schedule of variations: unix Running target unix Using /usr/share/dejagnu/baseboards/unix.exp as board description file for target. Using /usr/share/dejagnu/config/unix.exp as generic interface file for target. WARNING: Couldn't find tool config file for unix, using default. Running ../hammer.test/hammer.exp ... FAIL: test_hammer === hammer Summary === # of unexpected failures 1 WARNING: Couldn't find the global config file. WARNING: Couldn't find tool init file Test run by mockbuild on Wed Jun 22 14:49:05 2016 Native configuration is i686-pc-linux-gnu === spanner tests === Schedule of variations: unix Running target unix Using /usr/share/dejagnu/baseboards/unix.exp as board description file for target. Using /usr/share/dejagnu/config/unix.exp as generic interface file for target. WARNING: Couldn't find tool config file for unix, using default. Running ../spanner.test/spanner.exp ... === spanner Summary === # of expected passes 1 Makefile:467: recipe for target 'check-DEJAGNU' failed make[1]: *** [check-DEJAGNU] Error 1 make[1]: Leaving directory '/builddir/build/BUILD/automake-1.15/t/check12.dir/build' Makefile:817: recipe for target 'check-am' failed make: *** [check-am] Error 2 + test -f hammer.log + test -f hammer.sum + test -f spanner.log + test -f spanner.sum + grep 'FAIL: test_hammer' hammer.sum FAIL: test_hammer Koschei reports first failure with this build root change: java-1.8.0-openjdk-devel 1:1.8.0.77-1.b03.fc25 > 1:1.8.0.77-2.b03.fc25 git 2.8.0-1.fc25 > 2.8.1-1.fc25 java-1.8.0-openjdk-headless 1:1.8.0.77-1.b03.fc25 > 1:1.8.0.77-2.b03.fc25 java-1.8.0-openjdk 1:1.8.0.77-1.b03.fc25 > 1:1.8.0.77-2.b03.fc25 perl-Git 2.8.0-1.fc25 > 2.8.1-1.fc25 git-core 2.8.0-1.fc25 > 2.8.1-1.fc25 git-core-doc 2.8.0-1.fc25 > 2.8.1-1.fc25 harfbuzz 1.2.4-1.fc25 > 1.2.5-1.fc25 libproxy 0.4.12-2.fc25 > 0.4.12-3.fc25 libsolv 0.6.19-2.fc25 > 0.6.19-3.fc25 gzip 1.6-10.fc24 > 1.7-1.fc25 kernel-headers 4.6.0-0.rc1.git0.3.... > 4.6.0-0.rc2.git0.1....
Thanks for the report. Unfortunately, I was not correctly subscribed for Koschei notifications when the FTBFS occurred for the first time; so the analysis now: Something changed in dejagnu between 1.5 and 1.6, I'll have to look what exactly (my tip is that some return exit status changed in error case). What is weird is that the dejagnu is not mentioned in the list of changed packages (it would be good idea to keep the {build,root}.log files stored in koschei.
Created attachment 1173175 [details] Reproducer archive Reproducible by: $ tar xf TARBALL $ cd TARBALLDIR $ runtest HAMMER=./hammer SPANNER=./spanner --tool spanner --srcdir . Runtest returned '0' before, now it returns '2', which sounds like expected fix for runtest behavior. There seems to be two files missing in EXTRA_DIST :) so I'll fix the automake's testsuite.
Adding dejagnu maintainers to CC just to be sure that I'm here. Is this related to some known dejagnu bugfix?
s/I'm here/I'm not wrong here/
Fix proposed upstream: http://thread.gmane.org/gmane.comp.sysutils.automake.patches/9941 Build: http://koji.fedoraproject.org/koji/buildinfo?buildID=775943
Oh, OK, yet another testsuite failure, that might be the reason why testsuite started to fail for the first time in Koschei: ... PASS: t/distcheck-hook2.sh FAIL: t/distcheck-no-prefix-or-srcdir-override.sh PASS: t/distcheck-pr9579.sh ...
Backported one missing patch (so Koschei was right, update to gzip 1.7 originally caused this FTBFS): http://pkgs.fedoraproject.org/cgit/rpms/automake.git/commit/?id=c552a0666629018e39c531aeb