我已经设置了 MAAS 环境并使用 Juju 部署服务。我按照简单的 juju 教程尝试部署 Mysql 和 Wordpress。引导节点运行正常，没有任何错误。但是，在部署 Mysql（juju deploy --repository=<path> local:raring/mysql（使用 bzr 获取 mysql 和 wordpress charms））时，maas 节点使用 PXE 启动通过网络进行了 ubuntu 安装（所有这些似乎都正常工作）。
但是，charm 似乎没有出现。Juju 状态停留在待处理状态，日志（用于部署 charm 的 machine-1）似乎表明，start --system jujud-unit-mysql-0由于无法连接到系统总线 ( /var/run/dbus/system_bus_socket)，执行失败。
该论坛上还有其他帖子表明这可能是由/var/run和的问题引起的/run（但可能已在 raring 系列中修复），并且可能sudo需要使用命令。
显然，这是 Juju 内部产生的，我无法控制（使用 sudo 或不执行该命令来部署 charm）。有人可以帮忙调试这个问题吗？提前谢谢。
符咒状态
environment: maas
machines:
"0":
agent-state: started
agent-version: 1.16.0
dns-name: Honeybee-2.MaasCluster
instance-id: /MAAS/api/1.0/nodes/node-0abcb8ec-378b-11e3-b1d0-00155d964216/
series: raring
"1":
agent-state: started
agent-version: 1.16.0
dns-name: Honeybee-3.MaasCluster
instance-id: /MAAS/api/1.0/nodes/node-58eddf9c-379e-11e3-b1d0-00155d964216/
series: raring
services:
mysql:
charm: local:raring/mysql-309
exposed: false
relations:
cluster:
- mysql
units:
mysql/0:
agent-state: pending
machine: "1"
public-address: Honeybee-3.MaasCluster
#ssh [email protected]
# less /var/log/juju/machine-1.log
2013-10-23 20:32:05 INFO juju.worker.deployer deployer.go:146 deploying unit "mysql/0"
2013-10-23 20:32:05 INFO juju.worker.machiner machiner.go:52 "machine-1" started
2013-10-23 20:32:05 INFO juju.worker.upgrader upgrader.go:134 desired tool version: 1.16.0
2013-10-23 20:32:05 DEBUG juju.worker.deployer simple.go:91 state addresses: ["Honeybee-2.MaasCluster:37017"]
2013-10-23 20:32:05 DEBUG juju.worker.deployer simple.go:96 API addresses: ["Honeybee-2.MaasCluster:17070"]
2013-10-23 20:32:06 WARNING juju.worker.deployer simple.go:244 installer: cannot remove "/var/lib/juju/agents/unit-mysql-0": remove /var/lib/juju/agents/unit-mysql-0: directory not empty
2013-10-23 20:32:06 ERROR juju runner.go:211 worker: exited "deployer": exec ["start" "--system" "jujud-unit-mysql-0"]: exit status 1 (start: Unable to connect to system bus: Failed to connect to socket /var/run/dbus/system_bus_socket: No such file or directory)
2013-10-23 20:32:06 INFO juju runner.go:245 worker: restarting "deployer" in 3s
2013-10-23 20:32:09 INFO juju runner.go:253 worker: start "deployer"
2013-10-23 20:32:09 INFO juju.worker.deployer deployer.go:106 checking unit "mysql/0"
2013-10-23 20:32:09 INFO juju.worker.deployer deployer.go:106 checking unit "mysql/0"
(END)
#dpkg -l dbus
||/ Name Version Architecture Description
+++-==============-============-============-=================================
ii dbus 1.6.8-1ubunt amd64 simple interprocess messaging sys
答案1
为了添加有关此问题的更多信息：在基于 Saucy Salamander 的 MAAS 节点上部署“juju-gui”魅力时，我也遇到了同样的行为。
我也通过bzr（使用charm get juju-gui）获得了这个charm。
我注意到我们在不同的 charms、Ubuntu 系列甚至 Juju 版本上都遇到了同样的问题。
我的魔力状态：
services:
juju-gui:
charm: local:saucy/juju-gui-99
exposed: false
units:
juju-gui/0:
agent-state: pending
machine: "1"
public-address: node2.opc-ss.com
节点上的日志：
2013-11-19 08:51:08 INFO juju.worker.deployer deployer.go:146 deploying unit "juju-gui/0"
2013-11-19 08:51:08 INFO juju.worker.upgrader upgrader.go:134 desired tool version: 1.16.3
2013-11-19 08:51:08 INFO juju.worker.machiner machiner.go:52 "machine-1" started
2013-11-19 08:51:09 WARNING juju.worker.deployer simple.go:244 installer: cannot remove "/var/lib/juju/agents/unit-juju-gui-0": remove /var/lib/juju/agents/unit-juju-gui-0: directory not empty
2013-11-19 08:51:09 ERROR juju runner.go:211 worker: exited "deployer": exec ["start" "--system" "jujud-unit-juju-gui-0"]: exit status 1 (start: Unable to connect to system bus: Failed to connect to socket /var/run/dbus/system_bus_socket: No such file or directory)
2013-11-19 08:51:09 INFO juju runner.go:245 worker: restarting "deployer" in 3s
2013-11-19 08:51:09 INFO juju.provisioner provisioner_task.go:102 Starting up provisioner task machine-1
2013-11-19 08:51:12 INFO juju runner.go:253 worker: start "deployer"
2013-11-19 08:51:12 INFO juju.worker.deployer deployer.go:106 checking unit "juju-gui/0"
2013-11-19 08:51:12 INFO juju.worker.deployer deployer.go:106 checking unit "juju-gui/0"