What does your r10k.yaml look like for the postrun command? What you are
trying to do is possible of course. I have a postrun script that I call
that wraps and commands I want to perform post deploy.
Your Puppetfile should be in the environment/ folder. So
like /etc/puppetlabs/code/environments
So reading this right you want to have an agent run code out of the test
branch on the puppetserver and not the production branch? And it will work
fine if you change the codedir on the puppetserver?
The clients usually do not dictate the code directory on the remote
server. They tell the server
Hmm...
Do you have plugin sync turned on in the agent config? Should see
something like pluginsync = true in your puppet.conf. The error
specified seems to be having an issue fining the type which is sync'ed
from the master to the agents through plugin sync.
On Tue, Jul 1, 2014 at 1:50 PM, Pabl
I can weigh in a little on this...
As far as scalability of the puppet masters, it can be variable
depending on what kind of puppet code you are running. The largest
item that can ruin scalability is long catalog compilation time.
Another thing is run interval.
Current at my workplace we run a p
So how did you install Ruby 1.9.3? Did you create your own RPM or use
something else? The RPM packages for puppet will have dependencies
and then yum will try to satisfy them by looking in it's configured
yum repositories. It usually tries to find the latest version.
Looking at your output it lo
Some ruby gems are compiled and they need dependencies. In this case
you need the MySQL development files. Usually it is an RPM like
mysql-devel that you need in addition to gcc, make, etc...
The first error is probably because you specify versions in gem
differently than how you tried above. F
I would guess that the lens is error out when it attempts to load the
file. Have you tried to use augtool manually to see if you can make
your changes?
One thing I notice is that if a lens cannot parse a file it will say
in the error section. Taking a guess something like this:
# augtool
augtoo
Well the easiest way would be to make a puppet module and place your
facts in the correct path. Then enable pluginsync which causes the
fact to get downloaded to the puppet clients.
See: http://docs.puppetlabs.com/guides/plugins_in_modules.html
On Sun, Oct 13, 2013 at 3:00 AM, Sans wrote:
>
>
It looks like you do not have all the Ruby gems installed.
Storeconfigs requires Activerecord version 2.1 or later. On your
system depending on how you install your Ruby gems, you should run:
#see if Activerecord is installed
gem list
if you do not find it installed then you should install from
Ouch my eyes... Augeas can wreck your day sometimes. I have to ask
though, why not use the Puppetlabs Firewall Module? It's here:
https://forge.puppetlabs.com/puppetlabs/firewall
Then you would do something like this:
firewall { 'mayo_fw TCP/3':
ensure => present,
table => 'filter
Data like this should probably go into hiera. Then write modules to
use the data via parameterized classes or hiera() lookups. For
example we use the "datadir" directive in hiera to split out hiera
lookups to become environments aware. This would simplify your data
structure to where you are just
Not sure if you got help or not but that error is telling you that
filesys_group_acl is not a resource type. This is because you created
it as a class and not a defined resource type. Defined types use the
"define" keyword instead of the "class" keyword.
http://docs.puppetlabs.com/puppet/3/refer
Puppet uses Facter to determine the FQDN. I believe Facter uses a few
methods to try and figure out the domain part, namely:
hostname -f
dnsdomainname
parse out /etc/resolv.conf for the domain or search lines.
HTH,
Cory
On Tue, Sep 10, 2013 at 1:16 PM, Nathan Valentine wrote:
> Check both of t
Look at the apache::vhost definition. It looks to me that "template"
is no longer a parameter to the defined type. Since this looks like
puppetlabs-dashboard I would check out the puppetlabs-apache module to
find the apache::vhost defined type in
puppetlabs-apache/manifest/vhost.pp. If the param
We have lots of puppet clients on crappy bandwidth that would time out
like this as well. The option we changed to fix this is:
#Specify the timeout to wait for catalog in seconds
configtimeout = 600
The default time is like 60 or 120 secs. Another thing you should do
is check out the l
Hello:
I always seem to not be able to use localhost even with Posgtres bound
to all addresses. The command I tend to use is like:
psql -h 10.10.10.10 -d puppetdb -U puppetdb -W
then enter your password.
HTH
On Wed, Jul 24, 2013 at 2:01 PM, GregC wrote:
> I was able to run a psql puppetdb w
So that log looks like the Apache log... Do you have your
puppetmaster logging to syslog as well? There should be more than
just the "GET" lines from Apache. I think when you turn on --debug it
will increase the normal logging of the puppetmaster to syslog. That
is where you should see lines re
Not exactly sure if you tried this but I would added --debug to your
config.ru and then tail the logs on the puppetmaster. Clear out of
client SSL dir on the agent to start from scratch and make sure the
master does not have any leftover files from your client.
config.ru with debug:
$0 = "master
On Thu, Jul 18, 2013 at 1:53 PM, Jakov Sosic wrote:
> On 07/18/2013 08:29 PM, GregC wrote:
>
>> Load balancing will work without issue, there is a good suggestion on
>> using round robin DNS in Chapter 4 of Pro Puppet page 116.
>>
>
> Yeah but I am little suspicious about two masters sharing /var
Hello:
After upgrading to Puppet 3.2.3 I have noticed that the puppetmaster does
not log the catalog compilation time like it did in 3.1.1 or before. This
log line used to look like:
"myserver puppet-master[19002]: Compiled catalog for myhost in environment
production in 1.01 seconds"
Even if
20 matches
Mail list logo
