分类： 系统运维
2014-12-04 13:09:22
cat /tmp/puppetd.1415852379.out
info: Retrieving plugin
info: Caching catalog for sbx01o.XXX.com
info: Applying configuration version '1415852083'
notice: Finished catalog run in 0.12 seconds
怀疑是该repo没有sbx的class, 验证方法，登陆到nbt01o（puppet master)
[root@nbt01o ~]# /etc/puppet/external_node_v2.rb sbx01o.xxx.com
---
parameters:
callback_url: mcb-cn-mobile-qa.odd.blackspider.com
PORTAL_FQDN: cluster-cn-mobile-qa.odd.blackspider.com
hostgroup: SBX
puppetmaster: ""
replacement_repo:
ActiveEnvironments/THIRD_PARTY_RELEASE_TO_CHINA_QA_STABILISATION_LD
host_for_replacement_repo: packages
classes:{} # 为空，其他机器有。
------
解决办法：重新rebuild整个cluster,因为puppet master在nbt上，当nbt做完之后，每个机器定义的repo/class都确定了。
或在Foreman上：
Cofigure-->puppet 类-à从puppetmaster-proxy 导入,等待一会就有了.(未做这步之前,搜XXX_LD,结果为空)
