Amazon ec2 Chef尝试在EC2上部署Torquebox应用程序时失败

一般来说，我对Chef/ops很陌生，但我正在尝试在EC2上部署，我一直遇到这个错误。错误中提到的堆栈跟踪文件不存在，因此很难调试问题。此外，每当Chef失败时，我的EC2框上的权限就会被拒绝，我必须启动一个新实例。作为参考，此文件的原始来源如下：

错误如下所示：

  * script[install torquebox backstage] action run
    - execute "bash"  "/tmp/chef-script20131011-2067-1phzfkw"

[2013-10-11T03:47:51+00:00] ERROR: Running exception handlers
[2013-10-11T03:47:51+00:00] ERROR: Exception handlers complete
[2013-10-11T03:47:51+00:00] FATAL: Stacktrace dumped to /tmp/chef-solo/chef-stacktrace.out
Chef Client failed. 71 resources updated
[2013-10-11T03:47:51+00:00] FATAL: Chef::Exceptions::ChildConvergeError: Chef run process exited unsuccessfully (exit code 1)
ERROR: RuntimeError: chef-solo failed. See output above.

我的配方文件-torquebox.rb：

download_file = File.join("/tmp", File.basename(node.torquebox.download_url))

remote_file download_file do
  source  node.torquebox.download_url
  mode 00644
  action :create_if_missing
  notifies :run, "script[install-torquebox]", :immediately
end

script "install-torquebox" do
  Chef::Log.info("Installing torquebox zip file...")
  user 'torquebox'
  interpreter "bash"
  code <<-EOH
  unzip -o #{download_file} -d /opt/torquebox/
  ln -s /opt/torquebox/torquebox-2.3.2 /opt/torquebox/current
  EOH
  notifies :run, "execute[change ownership to torquebox]"
  not_if do
    File.exists? "/opt/torquebox/torquebox-2.3.2"
  end
  # not_if do
  #    File.exists? "/opt/torquebox/torquebox-2.2.0"
  #  end
  #action :nothing
end


template "/etc/profile.d/torquebox.sh" do
  source "torquebox.sh.erb"
  mode 00644
  owner 'root'
  variables(
    :jboss_user =>  node[:torquebox_env][:jboss_user],
    :torquebox_home => node[:torquebox_env][:home],
    :jboss_pidfile => node[:torquebox_env][:jboss_pidfile],
    :jboss_console_log => node[:torquebox_env][:jboss_console_log],
    :jboss_config => node[:torquebox][:configuration_file],
    :jruby_opts => node[:torquebox_env][:jruby_opts],
    :java_environment_file => node[:java][:java_environment]
  )
  notifies :restart, "service[jboss-as-standalone]"
end


# install torquebox backstage
#
execute "change ownership to torquebox" do
  user "root"
  cwd "/opt"
  Chef::Log.info("changing ownership for torquebox")
  command "chown -Rv 1000.1000 /opt/torquebox"
  notifies :run, "script[install torquebox backstage]"
  action :nothing
end

script "install torquebox backstage" do
  Chef::Log.info("Installing torquebox backstage file...")
  interpreter "bash"
  user "torquebox"
  cwd "/opt/torquebox"
  code <<-EOH
  export TORQUEBOX_HOME=/opt/torquebox/torquebox-2.3.2
  export JAVA_HOME=/opt/jdk7/
  export JBOSS_HOME=$TORQUEBOX_HOME/jboss
  export JRUBY_HOME=$TORQUEBOX_HOME/jruby
  export PATH=$JBOSS_HOME/bin:$JRUBY_HOME/bin:$JAVA_HOME/bin:$PATH
  EOH
  # jruby -S gem install torquebox-backstage
  # jruby -S gem install ruby-shadow
  # jruby -S backstage deploy
  not_if do
    File.exists? "/opt/torquebox/current/jruby/bin/backstage"
  end
end

directory '/opt/apps/' do
  owner "torquebox"
  group "torqubox"
  mode 00755
  action :create
  not_if do
    File.exists? "/opt/apps/"
  end
end

directory "/etc/jboss-as" do
  owner "root"
  group "root"
  mode 00755
  action :create
  not_if do
    File.exists? "/etc/jboss-as"
  end
end

directory "/var/log/jboss-as" do
  owner "torquebox"
  group "torquebox"
  mode 00755
  action :create
  not_if do
    File.exists? "/var/log/jboss-as"
  end
end

template "/etc/init.d/jboss-as-standalone" do
  source "jboss-as-standalone.sh.erb"
  variables(:environment_file => node[:torquebox][:environment_file])
  mode 00755
  owner 'root'
  notifies :restart, "service[jboss-as-standalone]"
end


service "jboss-as-standalone" do
  supports :status => true, :restart => true, :stop => true, :start => true
  action [ :enable, :start ]
end

#if clustered then if cluster name is staging_cluster, so proxy name will be staging_cluster_proxy
#so for every cookbook, it should be paired with two of them
proxy_nodes = []
clustered_nodes = []

Chef::Log.warn("clustered status #{node[:torquebox][:clustered]}")
if node[:torquebox][:clustered]
  clustered_nodes =  search(:node, "roles:#{node[:torquebox][:cluster_name]}")
  proxy_nodes = search(:node, "roles:#{node[:torquebox][:cluster_name]}_proxy")

  template "/opt/torquebox/current/jboss/standalone/configuration/standalone-ha.xml" do
    source "standalone-ha.xml.erb"
    variables(:node_name => node.name, :node_ipaddress => node.ipaddress, :cluster_name => node[:torquebox][:cluster_name], :clustered_nodes => clustered_nodes, :proxy_nodes => proxy_nodes )
    mode "0644"
    notifies :restart, "service[jboss-as-standalone]"
  end
end

if proxy_nodes.count == 0 and node[:torquebox][:clustered] == true
  Chef::Log.warn("There is no proxy defined, cluster may not function")
end

if !node[:torquebox][:clustered]
  template "/opt/torquebox/current/jboss/standalone/configuration/standalone.xml" do
    source "standalone.xml.erb"
    variables(:node_name => node.name, :node_ipaddress => node.ipaddress)
    mode "0644"
    notifies :restart, "service[jboss-as-standalone]"
  end
end

template "/opt/torquebox/current/jboss/bin/standalone.conf" do
  source "standalone.conf.erb"
  mode "00644"
  owner 'torquebox'
  variables(:jboss_config => node[:torquebox][:configuration_file])
  notifies :restart, "service[jboss-as-standalone]"
end

cookbook_file "/etc/jboss-as/jboss-as.conf" do
  source "jboss-as.conf"
  mode 00644
  owner 'root'
end


# NGINX!

# this enables our site, kinda like a2ensite
execute 'enable-site' do
  command "ln -sf /etc/nginx/sites-available/#{node[:server_name]} /etc/nginx/sites-enabled/#{node[:server_name]}"
  notifies :restart, 'service[nginx]'
end

# Our configuration template. Take a look at templates/nginx.conf.erb to see what's going on.
template "/etc/nginx/sites-available/#{node[:server_name]}" do
  source 'nginx.erb'
  owner 'root'
  group 'root'
  mode 0644
  notifies :run, "execute[enable-site]", :immediately
  variables(
     server_name: node.server_name
   )
end

service 'nginx'
service 'jboss-as-standalone'

download\u file=file.join（“/tmp”，file.basename（node.torquebox.download\u url））
远程\u文件下载\u文件吗
source node.torquebox.download\u url
模式00644
操作：如果缺少，则创建
通知：立即运行“脚本[install torquebox]”
结束
脚本“安装扭矩箱”执行以下操作
Chef:：Log.info（“安装torquebox zip文件…”）
用户“扭矩盒”
解释器“bash”
代码节点[：torquebox_env][：home]，
：jboss_pidfile=>node[：torquebox_env][：jboss_pidfile]，
：jboss_console_log=>node[：torquebox_env][：jboss_console_log]，
：jboss_config=>node[：torquebox][：configuration_file]，
：jruby\u opts=>node[：torquebox\u env][：jruby\u opts]，
：java_环境_文件=>node[：java][：java_环境]
)
通知：重新启动，“服务[jboss作为独立服务]”
结束
#在后台安装扭矩箱
#
执行“将所有权更改为扭矩箱”执行
用户“根”
cwd“/opt”
Chef:：Log.info（“更改torquebox的所有权”）
命令“chown-Rv 1000.1000/opt/torquebox”
通知：运行“脚本[install torquebox backstage]”
行动：没什么
结束
脚本“在后台安装torquebox”执行
Chef:：Log.info（“安装torquebox后台文件…”）
解释器“bash”
用户“扭矩盒”
cwd“/选择/扭矩箱”
代码true，：restart=>true，：stop=>true，：start=>true
操作[：启用，：启动]
结束
#如果集群，那么如果集群名称是staging\u cluster，那么代理名称将是staging\u cluster\u proxy
#因此，对于每一本食谱，它应该与其中两本配对
代理_节点=[]
集群_节点=[]
Chef:：Log.warn（“集群状态#{node[：torquebox][：集群]}”）
如果节点[：torquebox][：群集]
clustered_nodes=搜索（：node，“角色：{node[：torquebox][：cluster_name]}”）
proxy_nodes=search（：node，“角色：{node[：torquebox][：cluster_name]}\u proxy”）
模板“/opt/torquebox/current/jboss/standalone/configuration/standalone ha.xml”do
来源“standalone ha.xml.erb”
变量（：node_name=>node.name，：node_ipaddress=>node.ipaddress，：cluster_name=>node[：torquebox][：cluster_name]，：clustered_nodes=>clustered_nodes，：proxy_nodes=>proxy_nodes）
模式“0644”
通知：重新启动，“服务[jboss作为独立服务]”
结束
结束
如果proxy_nodes.count==0且节点[：torquebox][：clustered]==true
Chef:：Log.warn（“未定义代理，群集可能无法运行”）
结束
如果！节点[：扭矩盒][：群集]
模板“/opt/torquebox/current/jboss/standalone/configuration/standalone.xml”do
来源“standalone.xml.erb”
变量（：node\u name=>node.name，：node\u ipaddress=>node.ipaddress）
模式“0644”
通知：重新启动，“服务[jboss作为独立服务]”
结束
结束
模板“/opt/torquebox/current/jboss/bin/standalone.conf”do
来源“standalone.conf.erb”
模式“00644”
所有者“扭矩箱”
变量（：jboss_config=>node[：torquebox][：configuration_file]）
通知：重新启动，“服务[jboss作为独立服务]”
结束
cookbook_file“/etc/jboss as/jboss as.conf”do
来源“jbossas.conf”
模式00644
所有者“根”
结束
#NGINX！
#这使我们的网站，有点像一个敏感
执行“启用站点”do
命令“ln-sf/etc/nginx/sites available/#{node[：server\u name]}/etc/nginx/sites enabled/#{node[：server\u name]}”
通知：重新启动“服务[nginx]”
结束
#我们的配置模板。查看templates/nginx.conf.erb以了解发生了什么。
模板“/etc/nginx/sites available/#{node[：server_name]}”do
来源“nginx.erb”
所有者“根”
“根”组
模式0644
通知：运行，“执行[启用站点]”，：立即
变数(
服务器名称：node.server\u名称
)
结束
服务“nginx”
服务“jboss作为独立版”

/tmp/chef solo/chef stacktrace.out

应该在EC2实例上，而不是本地计算机上。检查一下。你能找到解决办法吗？如果你这样做了，你可以把它贴出来，并把你的答案标记为正确的？：）