Bug#923674: systemd: FTBFS (failing tests)

Michael Biebl biebl at debian.org
Sat Mar 9 21:43:49 GMT 2019


Am 09.03.19 um 22:27 schrieb Santiago Vila:
> On Sat, Mar 09, 2019 at 09:59:16PM +0100, Michael Biebl wrote:
>> Am 09.03.19 um 21:50 schrieb Michael Biebl:
>>> Am 09.03.19 um 21:00 schrieb Santiago Vila:
>>>> to which I also add these two lines:
>>> ..> /dev/urandom    /dev/random     none    rw,bind         0       0
>>>
>>> This might very well be the culprit, given that test-stat-util does the
>>> following check:
>>>
>>>         test_device_path_make_canonical_one("/dev/random");
>>>         test_device_path_make_canonical_one("/dev/urandom");
>>>
>>> https://github.com/systemd/systemd/blob/master/src/test/test-stat-util.c#L148
> 
> Good catch!
> 
>> Why are you bind mounting /dev/urandom to /dev/random (note the missing
>> 'u')?
> 
> It is a workaround for the fact that some packages FTBFS because of
> lack of entropy (mostly unit tests for cryptography-related packages).
> 
> Why are you testing /dev/random and /dev/urandom at all?
> 
> They are supposed to be sane in an autobuilder, no need to check them
> while building any package.

The point of the test is to check the implementation of the
device_path_make_{major_minor|canonical) functions provided by
libsystemd-shared.

For that it checks a set of device nodes, among others /dev/random and
/dev/urandom.

I'm having a hard timinig blaming the test here, tbh. Having
/dev/urandom provide a defined behaviour is something that can be
expected from a sanely configured system.

Regards,
Michael


-- 
Why is it that all of the instruments seeking intelligent life in the
universe are pointed away from Earth?

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
URL: <http://alioth-lists.debian.net/pipermail/pkg-systemd-maintainers/attachments/20190309/7f565343/attachment-0001.sig>


More information about the Pkg-systemd-maintainers mailing list