snapshot failed for timeout when getting image

[LiangChen]

When a docker container creating snapshot(use centos:centos7), get image failed for the reason of timeout.
But in fact the snapshot is successfully create( verify by "docker images" command).
The problem is repeatable and every time failed for timeout.

The reason is same with https://bugs.launchpad.net/nova-docker/+bug/1418803, but creating only failed for the first time,
and snapshot failed every time, so I think it is serious.

2015-04-06 22:27:44.153 4833 TRACE nova.compute.manager [instance: 24550e7d-7e0a-483f-8080-0c48bee77c80] Traceback (m
ost recent call last):
2015-04-06 22:27:44.153 4833 TRACE nova.compute.manager [instance: 24550e7d-7e0a-483f-8080-0c48bee77c80] File "/usr
/lib/python2.7/site-packages/nova/compute/manager.py", line 364, in decorated_function
2015-04-06 22:27:44.153 4833 TRACE nova.compute.manager [instance: 24550e7d-7e0a-483f-8080-0c48bee77c80] *args, *
*kwargs)
2015-04-06 22:27:44.153 4833 TRACE nova.compute.manager [instance: 24550e7d-7e0a-483f-8080-0c48bee77c80] File "/usr
/lib/python2.7/site-packages/nova/compute/manager.py", line 3021, in snapshot_instance
2015-04-06 22:27:44.153 4833 TRACE nova.compute.manager [instance: 24550e7d-7e0a-483f-8080-0c48bee77c80] task_sta
tes.IMAGE_SNAPSHOT)
2015-04-06 22:27:44.153 4833 TRACE nova.compute.manager [instance: 24550e7d-7e0a-483f-8080-0c48bee77c80] File "/usr
/lib/python2.7/site-packages/nova/compute/manager.py", line 3052, in _snapshot_instance
2015-04-06 22:27:44.153 4833 TRACE nova.compute.manager [instance: 24550e7d-7e0a-483f-8080-0c48bee77c80] update_t
ask_state)
2015-04-06 22:27:44.153 4833 TRACE nova.compute.manager [instance: 24550e7d-7e0a-483f-8080-0c48bee77c80] File "/usr
/lib/python2.7/site-packages/novadocker/virt/docker/driver.py", line 589, in snapshot
2015-04-06 22:27:44.153 4833 TRACE nova.compute.manager [instance: 24550e7d-7e0a-483f-8080-0c48bee77c80] instance
_id=instance['name'])
2015-04-06 22:27:44.153 4833 TRACE nova.compute.manager [instance: 24550e7d-7e0a-483f-8080-0c48bee77c80] NovaExceptio
n: Error saving image: UnixHTTPConnectionPool(host='localhost', port=None): Read timed out. (read timeout=10)
2015-04-06 22:27:44.153 4833 TRACE nova.compute.manager [instance: 24550e7d-7e0a-483f-8080-0c48bee77c80]

[Davanum Srinivas (DIMS) (dims-v)]

I'll bump up timeout in https://review.openstack.org/220610

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/220610
Committed: https://git.openstack.org/cgit/stackforge/nova-docker/commit/?id=523a1463e5f597f33bc5c6fde42ee064376d7469
Submitter: Jenkins
Branch: master

commit 523a1463e5f597f33bc5c6fde42ee064376d7469
Author: Davanum Srinivas <email address hidden>
Date: Fri Sep 4 11:27:02 2015 -0700

    Bump timeout and Docker API version

    Version 1.19 from the docker-py client
    https://github.com/docker/docker-py/blob/master/docker/constants.py#L1

    Our timeout is twice that of the default in docker-py

    Closes-Bug: #1440969
    Change-Id: I45427b6536586bdd2d154ec4e9634f7ce4e3c28d