Package: keyutils
Followup-For: Bug #757530
keyutils just tried to build on the official buildd:
https://buildd.debian.org/status/fetch.php?pkg=keyutils&arch=arm64&ver=1.5.9-5&stamp=1408551616
and there it behaves differently to the debian ports buildd:
dh_auto_test -- \
PATH=/«PKGBUILDDIR»:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games
\
LD_LIBRARY_PATH=/«PKGBUILDDIR» \
SKIPROOT=yes \
SKIPINSTALLED=yes \
MAXDESCLEN_NOCHECK=yes
make[2]: Entering directory '/«PKGBUILDDIR»'
grep: /etc/rpm/macros.dist: No such file or directory
make -C tests run
make[3]: Entering directory '/«PKGBUILDDIR»/tests'
sh runtest.sh keyctl/padd/noargs keyctl/padd/useradd keyctl/padd/bad-args
keyctl/clear/valid keyctl/clear/noargs keyctl/clear/bad-args keyctl/add/noargs
keyctl/add/useradd keyctl/add/bad-args keyctl/describing/valid
keyctl/describing/noargs keyctl/describing/bad-args keyctl/newring/valid
keyctl/newring/noargs keyctl/newring/bad-args keyctl/show/valid
keyctl/show/noargs keyctl/listing/valid keyctl/listing/noargs
keyctl/listing/bad-args keyctl/invalidate/valid keyctl/invalidate/noargs
keyctl/invalidate/bad-args keyctl/update/userupdate keyctl/update/noargs
keyctl/update/bad-args keyctl/search/valid keyctl/search/noargs
keyctl/search/bad-args keyctl/unlink/all keyctl/unlink/valid
keyctl/unlink/noargs keyctl/unlink/bad-args keyctl/instantiating/noargs
keyctl/instantiating/bad-args keyctl/link/recursion keyctl/link/valid
keyctl/link/noargs keyctl/link/bad-args keyctl/permitting/valid
keyctl/permitting/noargs keyctl/permitting/bad-args keyctl/requesting/valid
keyctl/requesting/piped keyct
l/requesting/noargs keyctl/requesting/bad-args keyctl/pupdate/userupdate
keyctl/pupdate/noargs keyctl/pupdate/bad-args keyctl/timeout/valid
keyctl/timeout/noargs keyctl/timeout/bad-args keyctl/reading/valid
keyctl/reading/noargs keyctl/reading/bad-args keyctl/revoke/valid
keyctl/revoke/noargs keyctl/revoke/bad-args keyctl/noargs keyctl/session/valid
keyctl/session/bad-args bugzillas/bz1033467
#### Some tests require root privileges.
#### It is recommended that this be run as root.
keyctl_describe: Function not implemented
Running with session keyring RHTS/keyctl/4475
keyctl_join_session_keyring: Function not implemented
Makefile:42: recipe for target 'run' failed
I'm not sure that I have permissions to run manual tests there to see if it's
reproducible.
On the debianports machine, where I do have permissions, which now has a 3.16
kernel, I can
reproduce the problem shown in my last response on this bug where it fails in:
+++ CHECK INVALID KEY TYPE
FAILED
which at least is consistent, but seems to be much earlier than the CHECK NO
LIST SESSION KEYRING in tests/keyctl/timeout/valid/ which we were previously
having anissue with.
tests/keyctl/padd/bad-args/test.out shows:
++++ BEGINNING TEST
+++ CHECK EMPTY KEY TYPE
echo -n stuff | keyctl padd wibble @p
add_key: Invalid argument
+++ CHECK UNSUPPORTED KEY TYPE
echo -n stuff | keyctl padd lizardsgizzards wibble @p
add_key: No such device
+++ CHECK INVALID KEY TYPE
echo -n stuff | keyctl padd .user wibble @p
add_key: No such device
=== FAILED ===
Session Keyring
967449289 --alswrv 1000 1000 keyring: RHTS/keyctl/6221
==============
+++ CHECK MAXLEN KEY TYPE
echo -n stuff | keyctl padd aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa wibble @p
add_key: No such device
+++ CHECK OVERLONG KEY TYPE
echo -n stuff | keyctl padd aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa wibble @p
add_key: Invalid argument
+++ CHECK ADD KEYRING WITH PAYLOAD
echo -n stuff | keyctl padd keyring wibble @p
add_key: Invalid argument
+++ CHECK MAXLEN DESC
echo -n stuff | keyctl padd user
aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa$
183194591
+++ CHECK BAD KEY ID
echo -n stuff | keyctl padd user wibble 0
add_key: Invalid argument
++++ FINISHED TEST: FAIL
Running the tests as root made no difference to the output (except
serial/session numbers, and to remove the 'Recommended to run these tests as
root' message.
I'll see if I can reproduce the build failure, but am dashing off to
debconf in little more than 24hrs so may not get time until after I
arrive (will you be there?)
Any ideas why the behaviour on the official debian buildd is quite
different might help us work out what's going on.
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
