I use XEN in RedHat linux (CentOS) for our QA environment. Within each XEN server I might have 20 – 30 servers. There are three XEN servers per QA environment, three QA environments. A constant problem is tracking where applications are installed within the VMs. I use puppet throughout the environments, including deployments of applications…
Continue ReadingPuppet
resolving yaml error in puppet
The error was: 246 – 9:08 info: Loading fact tmpfact err: Could not retrieve catalog: Could not parse YAML data for node app.net: syntax error on line 64, col 0: `time: 2011-07-31 01:52:21.901522 -04:00′ warning: Not using cache on failed catalog during a puppet run. To resolve, remove the .yaml files in /var/puppet/yaml/node and /var/puppet/yaml/facts…
Continue Readingtroubleshooting puppet variables
This is a practical use of searching for strings and finding and setting a missing puppet variable. Error from a puppet run: err: Could not retrieve catalog from remote server: Error 400 on SERVER: sysloghost variable must be defined in app_site module at /etc/puppet/environments/directory/modules/dstributed/app/manifests/log4j.pp:11 on node servernamed warning: Not using cache on failed catalog err:…
Continue Readingsearching for a string in a directory hierarchy
Much of the puppet infrastructure I work with is unmapped – it is the result of tickets submitted and variables asked for by developers or placed as solutions. This exists above the structure we designed and created intentionally. To discover where a variable is set in puppet, often I search through the entire puppet working…
Continue Readingpuppet errors: puppetlock and puppet certs
In working with puppet I often get the error “puppet run already in progress” when I know no such run is happening… The solution is to delete the file /var/lib/puppet/state/puppetdlock and then try the run again. The puppet command I use on the command line is: /usr/sbin/puppetd –server=puppet1 –onetime –no-daemonize –no-usecacheonfailure –ignorecache –no-noop -v…
Continue Reading