[systemd-devel] libsystemd-network tests failing in mock

Jan Synacek jsynacek at redhat.com
Fri Dec 12 06:11:14 PST 2014


Tom Gundersen <teg at jklm.no> writes:
> On Thu, Dec 11, 2014 at 9:14 AM, Jan Synáček <jsynacek at redhat.com> wrote:
>> test-dhcp-{client,server} are failing in mock:
>>
>> FAIL: test-dhcp-client
>> ======================
>> Assertion 'client' failed at ../src/libsystemd-network/sd-dhcp-client.c:138, function sd_dhcp_client_set_request_option(). Ignoring.
>> Assertion 'client' failed at ../src/libsystemd-network/sd-dhcp-client.c:169, function sd_dhcp_client_set_request_address(). Ignoring.
>> Assertion 'client' failed at ../src/libsystemd-network/sd-dhcp-client.c:182, function sd_dhcp_client_set_index(). Ignoring.
>> Assertion 'interface_index > 0' failed at ../src/libsystemd-network/sd-dhcp-client.c:185, function sd_dhcp_client_set_index(). Ignoring.
>> Assertion 'interface_index > 0' failed at ../src/libsystemd-network/sd-dhcp-client.c:185, function sd_dhcp_client_set_index(). Ignoring.
>> Assertion 'interface_index > 0' failed at ../src/libsystemd-network/sd-dhcp-client.c:185, function sd_dhcp_client_set_index(). Ignoring.
>> DHCP CLIENT (0x0): FREE
>> DHCP CLIENT (0xbe31128b): STARTED on ifindex 42
>> DHCP CLIENT (0xbe31128b): DISCOVER
>> DHCP CLIENT (0xbe31128b): STOPPED
>> DHCP CLIENT (0x0): FREE
>> DHCP CLIENT (0x45d3fd67): STARTED on ifindex 42
>> DHCP CLIENT (0x45d3fd67): DISCOVER
>> DHCP CLIENT (0x45d3fd67): OFFER
>> DHCP CLIENT (0x45d3fd67): REQUEST (requesting)
>> DHCP CLIENT (0x45d3fd67): ACK
>> DHCP CLIENT (0x45d3fd67): lease expires in 9min 57.698157s
>> DHCP CLIENT (0x45d3fd67): T2 expires in 8min 42.895614s
>> DHCP CLIENT (0x45d3fd67): T1 expires in 5min 101.292ms
>> DHCP CLIENT (0x45d3fd67): STOPPED: Operation not permitted
>
>
> This is really odd. I cannot reproduce, and looking at the code I
> cannot figure out how this could ever happen. I pushed a bit of extra
> logging just now, but even so I really don't see it.
>
> Any chance you can reproduce this locally and either let me know how,
> or go step through it with gdb and see where the EPERM originates?

Assuming that you run Fedora, you can patch the current specfile that is
in rawhide with the patch in [1]. Then, just execute "fedpkg
mockbuild". I tried to manually make the test fail directly inside the
mock buildroot, but couldn't make it work.  After patching the spec, you
can create an srpm with "fedpkg srpm" and then use mock direcly like so:

$ mock -r fedora-rawhide-x86_64 --no-cleanup --no-cleanup-after <path to srpm>

You'll then have the root and the results in /var/lib/mock/fedora-rawhide-x86_64/result.

[1] https://bugzilla.redhat.com/show_bug.cgi?id=1076119

Cheers,
-- 
Jan Synacek
Software Engineer, Red Hat
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 818 bytes
Desc: not available
URL: <http://lists.freedesktop.org/archives/systemd-devel/attachments/20141212/23de9e26/attachment.sig>


More information about the systemd-devel mailing list