Upgraded a legacy system's agent -- getting funky errors

[Peter Berghold]

Just after upgrading one of my legacy systems with a new puppet agent I noticed the agent wasn't running any more.

Running the agent in test mode I see:

Error: Could not create resources for managing Puppet's files and directories in sections [:main, :agent, :ssl]: undefined method `exists?' for #<Puppet::Type::Group:0x7fbdbf0a8410>

Error: Could not prepare for execution: Could not create resources for managing Puppet's files and directories in sections [:main, :agent, :ssl]: undefined method `exists?' for #<Puppet::Type::Group:0x7fbdbf0a8410>

undefined method `exists?' for #<Puppet::Type::Group:0x7fbdbf0a8410>

What the heck does that mean?  My puppet.conf file is short and sweet:

[agent]

        report = true

        configtimeout = 15000

Thoughts?  Should I just unmanage this host since it is getting retired this summer? 

Here are some facts from the system:

architecture => x86_64

augeasversion => 1.0.0

blockdevice_xvda1_size => 41875931136

blockdevice_xvda2_size => 1073741824

blockdevices => xvda1,xvda2

domain => chi.sharkrivertech.com

facterversion => 2.0.2

filesystems => ext3,iso9660

fqdn => hosting.chi.sharkrivertech.com

hardwareisa => x86_64

hardwaremodel => x86_64

hostname => hosting

id => root

interfaces => eth0,eth0_0,eth1,lo

ipaddress => 206.217.196.75

ipaddress_eth0 => 206.217.196.75

ipaddress_eth0_0 => 206.217.197.227

ipaddress_lo => 127.0.0.1

is_virtual => true

kernel => Linux

kernelmajversion => 2.6

kernelrelease => 2.6.32-71.el6.x86_64

kernelversion => 2.6.32

macaddress => <censored>

macaddress_eth0 =><censored>

macaddress_eth0_0 => <censored>

macaddress_eth1 => <censored>

memoryfree => 1005.02 MB

memoryfree_mb => 1005.02

memorysize => 1.95 GB

memorysize_mb => 1999.39

mtu_eth0 => 1500

mtu_eth0_0 => 1500

mtu_eth1 => 1500

mtu_lo => 16436

netmask => 255.255.255.0

netmask_eth0 => 255.255.255.0

netmask_eth0_0 => 255.255.255.0

netmask_lo => 255.0.0.0

network_eth0 => <censored>

network_eth0_0 => <censored>

network_lo => 127.0.0.0

operatingsystem => CentOS

operatingsystemmajrelease => 6

operatingsystemrelease => 6.2

osfamily => RedHat

path => /usr/local/sbin:/usr/local/bin:/sbin:/bin:/usr/sbin:/usr/bin:/root/bin

physicalprocessorcount => 1

processor0 => Intel(R) Xeon(R) CPU           E5520  @ 2.27GHz

processor1 => Intel(R) Xeon(R) CPU           E5520  @ 2.27GHz

processorcount => 2

ps => ps -ef

puppetversion => 3.6.2

rubysitedir => /usr/lib/ruby/site_ruby/1.8

rubyversion => 1.8.7

selinux => false

--

Peter L. Berghold                       Salty....@gmail.com

http://blog.berghold.net

[jcbollinger]

On Friday, June 20, 2014 9:44:59 AM UTC-5, Salty Old Cowdawg wrote:

Just after upgrading one of my legacy systems with a new puppet agent I noticed the agent wasn't running any more.

Running the agent in test mode I see:

Error: Could not create resources for managing Puppet's files and directories in sections [:main, :agent, :ssl]: undefined method `exists?' for #<Puppet::Type::Group:0x7fbdbf0a8410>

Error: Could not prepare for execution: Could not create resources for managing Puppet's files and directories in sections [:main, :agent, :ssl]: undefined method `exists?' for #<Puppet::Type::Group:0x7fbdbf0a8410>

undefined method `exists?' for #<Puppet::Type::Group:0x7fbdbf0a8410>

What the heck does that mean?  My puppet.conf file is short and sweet:

[agent]

        report = true

        configtimeout = 15000

Thoughts?  Should I just unmanage this host since it is getting retired this summer? 

It looks like you have an incompatible mix of Puppet bits on your system.  That likely results either from the old version not being completely removed during the upgrade, or from there always having been multiple copies installed (but that not previously having been a problem).

Supposing that you are using Puppet RPMs, check for a Puppet gem, and for a local from-source Puppet installation.  RPM can help you identify files that don't belong to the Puppet rpm.

You'll have to judge for yourself whether the work (probably not too much) is worth it for this system.


John

[Peter Berghold]

On Mon, Jun 23, 2014 at 9:24 AM, jcbollinger <John.Bo...@stjude.org> wrote:

On Friday, June 20, 2014 9:44:59 AM UTC-5, Salty Old Cowdawg wrote:

You'll have to judge for yourself whether the work (probably not too much) is worth it for this system.

Considering my timeframe for decomming the machine is end of August, that's a very valid point.  It is the last of my CenOS 5 machines left over from when I had my own hosting company and the last "hosting customer" has agreed to be migrated.   (Hosting customer:  Non-Profit I donated the space to)