我正在尝试在 AWS 上使用 Deis 推出一个 Node 应用程序（该应用程序已经为我工作了几个月，但最近才停止工作）。我现在注意到，当我推送我的应用程序时，并没有创建一个新版本。但是，如果我使用 deis config:set 更改环境变量，则会创建一个新版本；但是，它没有与之关联的构建（尝试扩展流程时出现以下错误：“没有与此版本关联的构建”）。
推送应用程序时，一切似乎都很顺利，我看到：
Trigger 0, RUN mkdir -p /app
Step 0 : RUN mkdir -p /app
---> Using cache
Trigger 1, WORKDIR /app
Step 0 : WORKDIR /app
---> Using cache
Trigger 2, ADD slug.tgz /app
Step 0 : ADD slug.tgz /app
---> 27a0d5cb7ff0
Removing intermediate container e4345a7b152a
Step 1 : ENV GIT_SHA f959d84d4496c48e39b045d727527d203d4e17a3
---> Running in 32ed021a1456
---> 03e742bd6cf4
Removing intermediate container 32ed021a1456
Successfully built 03e742bd6cf4
-----> Pushing image to private registry
查看注册表的日志（这似乎是导致问题的原因）时，我看到：
Nov 25 05:12:43 ip-10-0-10-252.ec2.internal sh[12257]: Traceback (most recent call last):
Nov 25 05:12:43 ip-10-0-10-252.ec2.internal sh[12257]: File "/app/bin/create_bucket", line 23, in <module>
Nov 25 05:12:43 ip-10-0-10-252.ec2.internal sh[12257]: if name not in (bucket.name for bucket in conn.get_all_buckets()):
Nov 25 05:12:43 ip-10-0-10-252.ec2.internal sh[12257]: File "/usr/lib/python2.7/site-packages/boto/s3/connection.py", line 436, in get_all_buckets
Nov 25 05:12:43 ip-10-0-10-252.ec2.internal sh[12257]: response = self.make_request('GET', headers=headers)
Nov 25 05:12:43 ip-10-0-10-252.ec2.internal sh[12257]: File "/usr/lib/python2.7/site-packages/boto/s3/connection.py", line 664, in make_request
Nov 25 05:12:43 ip-10-0-10-252.ec2.internal sh[12257]: retry_handler=retry_handler
Nov 25 05:12:43 ip-10-0-10-252.ec2.internal sh[12257]: File "/usr/lib/python2.7/site-packages/boto/connection.py", line 1068, in make_request
Nov 25 05:12:43 ip-10-0-10-252.ec2.internal sh[12257]: retry_handler=retry_handler)
Nov 25 05:12:43 ip-10-0-10-252.ec2.internal sh[12257]: File "/usr/lib/python2.7/site-packages/boto/connection.py", line 943, in _mexe
Nov 25 05:12:43 ip-10-0-10-252.ec2.internal sh[12257]: response = connection.getresponse()
Nov 25 05:12:43 ip-10-0-10-252.ec2.internal sh[12257]: File "/usr/lib/python2.7/httplib.py", line 1073, in getresponse
Nov 25 05:12:43 ip-10-0-10-252.ec2.internal sh[12257]: response.begin()
Nov 25 05:12:43 ip-10-0-10-252.ec2.internal sh[12257]: File "/usr/lib/python2.7/httplib.py", line 415, in begin
Nov 25 05:12:43 ip-10-0-10-252.ec2.internal sh[12257]: version, status, reason = self._read_status()
Nov 25 05:12:43 ip-10-0-10-252.ec2.internal sh[12257]: File "/usr/lib/python2.7/httplib.py", line 371, in _read_status
Nov 25 05:12:43 ip-10-0-10-252.ec2.internal sh[12257]: line = self.fp.readline(_MAXLINE + 1)
Nov 25 05:12:43 ip-10-0-10-252.ec2.internal sh[12257]: File "/usr/lib/python2.7/socket.py", line 476, in readline
Nov 25 05:12:43 ip-10-0-10-252.ec2.internal sh[12257]: data = self._sock.recv(self._rbufsize)
Nov 25 05:12:43 ip-10-0-10-252.ec2.internal sh[12257]: socket.timeout: timed out
Nov 25 05:12:44 ip-10-0-10-252.ec2.internal systemd[1]: deis-registry@1.service: Main process exited, code=exited, status=1/FAILURE
Nov 25 05:12:44 ip-10-0-10-252.ec2.internal docker[17529]: Error response from daemon: no such id: deis-registry
Nov 25 05:12:44 ip-10-0-10-252.ec2.internal docker[17529]: Error: failed to stop containers: [deis-registry]
Nov 25 05:12:44 ip-10-0-10-252.ec2.internal systemd[1]: deis-registry@1.service: Unit entered failed state.
Nov 25 05:12:44 ip-10-0-10-252.ec2.internal systemd[1]: deis-registry@1.service: Failed with result 'exit-code'.
Nov 25 05:12:49 ip-10-0-10-252.ec2.internal systemd[1]: deis-registry@1.service: Service hold-off time over, scheduling restart.
Nov 25 05:12:49 ip-10-0-10-252.ec2.internal systemd[1]: Starting deis-registry...
Nov 25 05:12:49 ip-10-0-10-252.ec2.internal etcdctl[17601]: Error: 105: Key already exists (/deis/cache) [18268602]
Nov 25 05:12:49 ip-10-0-10-252.ec2.internal systemd[1]: Started deis-registry.
Nov 25 05:21:55 ip-10-0-10-252.ec2.internal sh[17637]: Traceback (most recent call last):
Nov 25 05:21:55 ip-10-0-10-252.ec2.internal sh[17637]: File "/app/bin/create_bucket", line 23, in <module>
Nov 25 05:21:55 ip-10-0-10-252.ec2.internal sh[17637]: if name not in (bucket.name for bucket in conn.get_all_buckets()):
Nov 25 05:21:55 ip-10-0-10-252.ec2.internal sh[17637]: File "/usr/lib/python2.7/site-packages/boto/s3/connection.py", line 436, in get_all_buckets
Nov 25 05:21:55 ip-10-0-10-252.ec2.internal sh[17637]: response = self.make_request('GET', headers=headers)
Nov 25 05:21:55 ip-10-0-10-252.ec2.internal sh[17637]: File "/usr/lib/python2.7/site-packages/boto/s3/connection.py", line 664, in make_request
Nov 25 05:21:55 ip-10-0-10-252.ec2.internal sh[17637]: retry_handler=retry_handler
Nov 25 05:21:55 ip-10-0-10-252.ec2.internal sh[17637]: File "/usr/lib/python2.7/site-packages/boto/connection.py", line 1068, in make_request
Nov 25 05:21:55 ip-10-0-10-252.ec2.internal sh[17637]: retry_handler=retry_handler)
Nov 25 05:21:55 ip-10-0-10-252.ec2.internal sh[17637]: File "/usr/lib/python2.7/site-packages/boto/connection.py", line 943, in _mexe
Nov 25 05:21:55 ip-10-0-10-252.ec2.internal sh[17637]: response = connection.getresponse()
Nov 25 05:21:55 ip-10-0-10-252.ec2.internal sh[17637]: File "/usr/lib/python2.7/httplib.py", line 1073, in getresponse
Nov 25 05:21:55 ip-10-0-10-252.ec2.internal sh[17637]: response.begin()
Nov 25 05:21:55 ip-10-0-10-252.ec2.internal sh[17637]: File "/usr/lib/python2.7/httplib.py", line 415, in begin
Nov 25 05:21:55 ip-10-0-10-252.ec2.internal sh[17637]: version, status, reason = self._read_status()
Nov 25 05:21:55 ip-10-0-10-252.ec2.internal sh[17637]: File "/usr/lib/python2.7/httplib.py", line 371, in _read_status
Nov 25 05:21:55 ip-10-0-10-252.ec2.internal sh[17637]: line = self.fp.readline(_MAXLINE + 1)
Nov 25 05:21:55 ip-10-0-10-252.ec2.internal sh[17637]: File "/usr/lib/python2.7/socket.py", line 476, in readline
Nov 25 05:21:55 ip-10-0-10-252.ec2.internal sh[17637]: data = self._sock.recv(self._rbufsize)
Nov 25 05:21:55 ip-10-0-10-252.ec2.internal sh[17637]: socket.timeout: timed out
Nov 25 05:21:55 ip-10-0-10-252.ec2.internal systemd[1]: deis-registry@1.service: Main process exited, code=exited, status=1/FAILURE
Nov 25 05:21:55 ip-10-0-10-252.ec2.internal docker[22783]: Error response from daemon: no such id: deis-registry
Nov 25 05:21:55 ip-10-0-10-252.ec2.internal docker[22783]: Error: failed to stop containers: [deis-registry]
Nov 25 05:21:55 ip-10-0-10-252.ec2.internal systemd[1]: deis-registry@1.service: Unit entered failed state.
Nov 25 05:21:55 ip-10-0-10-252.ec2.internal systemd[1]: deis-registry@1.service: Failed with result 'exit-code'.
Nov 25 05:22:00 ip-10-0-10-252.ec2.internal systemd[1]: deis-registry@1.service: Service hold-off time over, scheduling restart.
Nov 25 05:22:00 ip-10-0-10-252.ec2.internal systemd[1]: Starting deis-registry...
Nov 25 05:22:00 ip-10-0-10-252.ec2.internal etcdctl[22865]: Error: 105: Key already exists (/deis/cache) [18274121]
Nov 25 05:22:00 ip-10-0-10-252.ec2.internal systemd[1]: Started deis-registry.
在查看构建器时，它似乎也可以工作，因为它说它已部署：
Nov 25 05:39:27 ip-10-0-10-252.ec2.internal sh[3142]: Waiting for git-receive to run.
Nov 25 05:39:27 ip-10-0-10-252.ec2.internal sh[3142]: Waiting for deploy.
Nov 25 05:40:17 ip-10-0-10-252.ec2.internal sh[3142]: [info] Deploy complete.
（注意：时间戳在上面的例子中是不同的，因为我稍后在不同的推送尝试中添加了这个）
最后，当我运行 deisctl list 时，所有内容都显示为活动并正在运行，因此看起来没有任何问题。
似乎很多网上的人都遇到了这个问题，但我还没有看到任何可以解决这个问题的明确回应。
我尝试重新启动注册表以及销毁整个应用程序并重新启动它。
如果它在上下文方面有帮助，那么当我注意到我的 /var/lib/docker 的使用率为 99% 时，这个问题就开始发生了——我按照 Deis 文档的指示清理了图像，所以我现在的使用率为 60%；但是，我仍然遇到完全相同的问题。
谢谢！