Heat fedora image download times out job

[Matthew Treinish]

On stable/icehouse jobs it looks like images used by devstack aren't be properly cached and it's going out to the network to grab these images. This causes occasional timeouts killing the jobs, for example:

http://logs.openstack.org/45/129045/1/gate/gate-tempest-dsvm-neutron-heat-slow/7b888df/logs/devstacklog.txt.gz#_2014-10-21_21_20_56_436

My guess is that we aren't caching the images from the stable devstack repos just on master. Which means we will have this issue when we change the image version on the master branch.

[Joe Gordon (jogo)]

Looks like we should backport the following changes:

https://review.openstack.org/#/c/88481/
https://review.openstack.org/#/c/102793/

[OpenStack Infra (hudson-openstack)]

Related fix proposed to branch: master
Review: https://review.openstack.org/130402

[OpenStack Infra (hudson-openstack)]

Change abandoned by Joe Gordon (<email address hidden>) on branch: master
Review: https://review.openstack.org/130402
Reason: ran things locally instead

[Joe Gordon (jogo)]

Clark, Jeremy and myself looked into this one a bit. When running the scripts locally everything worked. But the nodepool logs tell a different story. Not sure why this is working locally and not on nodepool

[Matthew Treinish (treinish)]

The job which was triggering this issue was removed with:

https://review.openstack.org/#/c/130633/