[openstack-dev] sqlalchemy-utils fails devstack install
Beliveau, Ludovic
Ludovic.Beliveau at windriver.com
Thu Feb 25 15:17:31 UTC 2016
A bug has been raised against SQLAlchemy-Utils and a fixed has been proposed (but not yet merged):
https://github.com/kvesteri/sqlalchemy-utils/pull/193
Workaround for running devstack:
- Download SQLAlchemy-Utils source code (see https://sqlalchemy-utils.readthedocs.org/en/latest/installation.html)
- Patch source code locally based on the proposed fix
- Install the package
Regards,
/ludovic
On 02/25/2016 08:27 AM, Sean Dague wrote:
> On 02/24/2016 11:45 PM, Watanabe, Isao wrote:
>> Hello team
>>
>> Does anyone know about why sqlalchemy-utils fails devstack install since about 3:00 UTC, Feb 25th?
>>
>> Our CI start to fail and in log it says:
>>
>> 2016-02-25 03:34:22.515 | Collecting SQLAlchemy-Utils===0.31.6 (from -c /opt/stack/new/requirements/upper-constraints.txt (line 24))
>> 2016-02-25 03:34:22.602 | Downloading SQLAlchemy-Utils-0.31.6.tar.gz (112kB)
>> 2016-02-25 03:34:23.031 | Complete output from command python setup.py egg_info:
>> 2016-02-25 03:34:23.031 | error in SQLAlchemy-Utils setup command: 'extras_require' must be a dictionary whose values are strings or lists of strings containing valid project/version requirement specifiers.
>> 2016-02-25 03:34:23.031 |
>> 2016-02-25 03:34:23.031 | ----------------------------------------
>> 2016-02-25 03:34:23.062 | Command "python setup.py egg_info" failed with error code 1 in /tmp/pip-build-eRYND2/SQLAlchemy-Utils
>>
>> However, in mirror[1] which we are using, the package is just there, which confused me.
>> [1] http://mirror.dfw.rax.openstack.org/pypi/simple/sqlalchemy-utils/
>>
>> Thanks for any help.
>>
>> Best regards,
>> Watanabe.isao
> The failure only exists if you are installing from tar.gz and not from
> wheels. For performance reasons the upstream gate builds a wheel mirror
> and installs from that.
>
> Those wheels were built on old setuptools, work on any version. There is
> a setuptools bug registered for this, which should hopefully be fixed today.
>
> -Sean
>
More information about the OpenStack-dev
mailing list