A bunch of Travis errors are being caused (apparently) by travis-wheels.scikit-image.org not responding. I haven't been following the discussion closely, but my question is, do we need to get rid of it for the Travis build so that we can have a stable build system at the sprint? Or do we perhaps want to add a condition to use regular pip builds if travis-wheels is unavailable?
On 12/07/2014 05:00 PM, Juan Nunez-Iglesias wrote:
A bunch of Travis errors are being caused (apparently) by travis-wheels.scikit-image.org not responding. I haven't been following the discussion closely, but my question is, do we need to get rid of it for the Travis build so that we can have a stable build system at the sprint? Or do we perhaps want to add a condition to use regular pip builds if travis-wheels is unavailable?
-- You received this message because you are subscribed to the Google Groups "scikit-image" group. To unsubscribe from this group and stop receiving emails from it, send an email to scikit-image+unsubscribe@googlegroups.com <mailto:scikit-image+unsubscribe@googlegroups.com>. For more options, visit https://groups.google.com/d/optout.
We know have http://wheels.scipy.org/ which has OS-X and Linux wheels for many SciPy stack modules. http://travis-wheels.scikit-image.org/ still responds for me and seems to have more files than the scipy.org domain. Github issue #1057 is where we are discussing the wheels, Windows CI, and related topic, https://github.com/scikit-image/scikit-image/pull/1057. If there are files you need at either site I can add them easily. Cheers, - Jonathan Helmus
Hey Jonathan, I'm not suggesting that travis-wheels is unresponsive all the time or missing files. I'm just saying that many (~30% guesstimate) builds have been failing because of a travis-wheels.scikit-image.org error. There is clearly a problem with uptime or with response time. I don't have any files that I need. Steven is the architect of the present build system (thanks Steven! =) so he'll know best how to deal with this. Juan. On Tue, Dec 9, 2014 at 3:38 AM, Jonathan Helmus <jjhelmus@gmail.com> wrote:
On 12/07/2014 05:00 PM, Juan Nunez-Iglesias wrote:
A bunch of Travis errors are being caused (apparently) by travis-wheels.scikit-image.org not responding. I haven't been following the discussion closely, but my question is, do we need to get rid of it for the Travis build so that we can have a stable build system at the sprint? Or do we perhaps want to add a condition to use regular pip builds if travis-wheels is unavailable?
-- You received this message because you are subscribed to the Google Groups "scikit-image" group. To unsubscribe from this group and stop receiving emails from it, send an email to scikit-image+unsubscribe@googlegroups.com <mailto:scikit-image+unsubscribe@googlegroups.com>. For more options, visit https://groups.google.com/d/optout. We know have http://wheels.scipy.org/ which has OS-X and Linux wheels for many SciPy stack modules. http://travis-wheels.scikit-image.org/ still responds for me and seems to have more files than the scipy.org domain. Github issue #1057 is where we are discussing the wheels, Windows CI, and related topic, https://github.com/scikit-image/scikit-image/pull/1057. If there are files you need at either site I can add them easily. Cheers, - Jonathan Helmus -- You received this message because you are subscribed to the Google Groups "scikit-image" group. To unsubscribe from this group and stop receiving emails from it, send an email to scikit-image+unsubscribe@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Juan, I'm not sure what is causing the recent failures. We have been using the new travis-wheels site for over a month: https://github.com/scikit-image/scikit-image/commit/b6f775a7cf8c775d89e8fe20.... Perhaps the rackspace server is having some trouble. - Steve On Sunday, December 7, 2014 5:00:58 PM UTC-6, Juan Nunez-Iglesias wrote:
A bunch of Travis errors are being caused (apparently) by travis-wheels.scikit-image.org not responding. I haven't been following the discussion closely, but my question is, do we need to get rid of it for the Travis build so that we can have a stable build system at the sprint? Or do we perhaps want to add a condition to use regular pip builds if travis-wheels is unavailable?
Ok no worries -- I just want to make sure that I'm not way off on what the causes of the failures are... Sometimes Travis errors are misleading. Who's running the actual server and do we have any logs about uptime? On Tue, Dec 9, 2014 at 1:41 PM, Steven Silvester <steven.silvester@gmail.com
wrote:
Juan,
I'm not sure what is causing the recent failures. We have been using the new travis-wheels site for over a month: https://github.com/scikit-image/scikit-image/commit/b6f775a7cf8c775d89e8fe20.... Perhaps the rackspace server is having some trouble.
- Steve
On Sunday, December 7, 2014 5:00:58 PM UTC-6, Juan Nunez-Iglesias wrote:
A bunch of Travis errors are being caused (apparently) by travis-wheels.scikit-image.org not responding. I haven't been following the discussion closely, but my question is, do we need to get rid of it for the Travis build so that we can have a stable build system at the sprint? Or do we perhaps want to add a condition to use regular pip builds if travis-wheels is unavailable?
--
You received this message because you are subscribed to the Google Groups "scikit-image" group. To unsubscribe from this group and stop receiving emails from it, send an email to scikit-image+unsubscribe@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Thanks Oliver, implemented. On Sunday, December 7, 2014 5:00:58 PM UTC-6, Juan Nunez-Iglesias wrote:
A bunch of Travis errors are being caused (apparently) by travis-wheels.scikit-image.org not responding. I haven't been following the discussion closely, but my question is, do we need to get rid of it for the Travis build so that we can have a stable build system at the sprint? Or do we perhaps want to add a condition to use regular pip builds if travis-wheels is unavailable?
participants (3)
-
Jonathan Helmus -
Juan Nunez-Iglesias -
Steven Silvester