问题1: luci服务无法启动

 

shell>service luci start

Unable to create the luci base configuration file

(`/var/lib/luci/etc/luci.ini').

Start luci...                                             [FAILED]

提示无法创建luci基础配置文件/var/lib/luci/etc/luci.ini.

手动建立/var/lib/luci/etc/luci.ini:

touch /var/lib/luci/etc/luci.ini

 

建立好之后,重新启动luci服务:

shell>service luci start

Unable to create the luci database file(`/var/lib/luci/data/luci.db').

Start luci...                                              [FAILED]

提示无法建立luci数据库文件/var/lib/luci/data/luci.db.

手动建立/var/lib/luci/data/luci.db:

touch /var/lib/luci/data/luci.db

建立好之后,重新启动luci服务:

shell>service luci start

luci服务好像没有报错了.

查看luci状态:

shell>service luci status

没有任何输出

 

查看日志:

shell>tail -n 200 /var/log/luci/luci.log

Traceback (most recent call last):

  File"/usr/bin/paster", line 9, in

   load_entry_point('PasteScript==1.7.3', 'console_scripts', 'paster')()

  File"/usr/lib/python2.6/site-packages/paste/script/command.py", line 84,in run

   invoke(command, command_name, options, args[1:])

  File"/usr/lib/python2.6/site-packages/paste/script/command.py", line 123,in invoke

    exit_code =runner.run(args)

  File"/usr/lib/python2.6/site-packages/paste/script/command.py", line 218,in run

    result =self.command()

  File"/usr/lib/python2.6/site-packages/paste/script/serve.py", line 274,in command

   relative_to=base, global_conf=vars)

  File"/usr/lib/python2.6/site-packages/paste/script/serve.py", line 308,in loadserver

    relative_to=relative_to,**kw)

  File"/usr/lib/python2.6/site-packages/paste/deploy/loadwsgi.py", line210, in loadserver

    returnloadobj(SERVER, uri, name=name, **kw)

  File"/usr/lib/python2.6/site-packages/paste/deploy/loadwsgi.py", line224, in loadobj

   global_conf=global_conf)

  File"/usr/lib/python2.6/site-packages/paste/deploy/loadwsgi.py", line248, in loadcontext

   global_conf=global_conf)

  File"/usr/lib/python2.6/site-packages/paste/deploy/loadwsgi.py", line278, in _loadconfig

    return loader.get_context(object_type,name, global_conf)

  File"/usr/lib/python2.6/site-packages/paste/deploy/loadwsgi.py", line363, in get_context

    object_type,name=name)

  File"/usr/lib/python2.6/site-packages/paste/deploy/loadwsgi.py", line528, in find_config_section

   self.filename))

LookupError: No section 'init' (prefixed by 'server')found in config /var/lib/luci/etc/luci.ini

Removing PID file /var/run/luci/luci.pid

原来luci服务启动失败了.

 

问题原因:

安装luci软件包时是使用的yum,而yum源中包含了epel源,在安装过程中使用了epel源里的包.而epel源里的包是最新版本的,与luci有些包不兼容.

 

解决办法:

shell>yum erase jabberpy.noarchpython-repoze-who-friendlyform.noarch python-tw-forms.noarch

shell>yum install luci --disablerepo=*epel*

 

英文解释:

When external yum repository is needed and therespective packages tend

to supersede the distribution's ones -- as justobserved with EPEL offering

newer python-repoze-who-friendlyform andpython-tw-forms -- it is advisable

to edit respective repository definition (e.g.,/etc/yum.repos.d/epel.repo

added with installation of epel-release package) andexclude colliding

packages.

 

In this case, line like this might help (be aware of"python-*" impact):

---

exclude=python-* TurboGears2 pyOpenSSL

 

 

问题2: 登录https://hostname:8084 用户名和密码

解决办法: 本机用户和密码, 可以使用root和root的密码

注: centos 5和rhel 5中的redhat cluster是使用luci_admin init命令给admin用户初始化密码,然后启动luci服务,在浏览器中使用admin和设置的密码登录.

 

问题3: 添加集群时: Unable to establish anSSL connection to server1.com:11111 after 5 tries.

解决办法:

启动ricci服务:

shell>service ricci start

 

问题4: 添加节点时: Authentication to thericci agent at 192.168.1.3.:11111 failed

解决: 给ricci添加密码

shell>passwd ricci

 

问题5: cman启动时提示Cannot find nodename in cluster.conf

[root@mfsmaster ~]# service cman start

Starting cluster:

   Checking ifcluster has been disabled at boot...       [  OK  ]

   CheckingNetwork Manager...                            [  OK  ]

   Globalsetup...                                        [  OK  ]

   Loadingkernel modules...                               [  OK  ]

   Mountingconfigfs...                                   [  OK  ]

   Startingcman... Cannot find node name in cluster.conf

Unable to get the configuration

Cannot find node name in cluster.conf

cman_tool: corosync daemon didn't start Check clusterlogs for details

                                                          [FAILED]

解决办法:

当前hostname不是cluster.conf中所定义的,修改hostname为cluster.conf中定义的

假设cluster.conf定义的节点域名是server1.com,则设置域名:

hostname server1.com

修改/etc/sysconfig/network

HOSTNAME=server1.com