Hi Steve,
All custom facts are delivered to all nodes within a puppetmaster's
control. IOW, having a module that has a custom fact will deploy that
fact to every node, regardless of if that node uses, or is aware of the
fact, or module in question.
It is generally expected that the fact
Hi David,
I've seen a setup that has an exec which fires apt-get update. This exec
has refreshonly = true. All Apt::Source resources are instantiated via a
profile class kicking off create_resources. All Apt::Source resources
notify the apt-get update exec.
I think something similar to that
Hi Guys,
I have a few questions about 3.6 directory environments, which we're
looking to adopt.
Currently the most pressing surrounds the integration of r10k and hiera…
I believe I want to store hieradata inside the r10k repos, so that each
r10k repo (I'm planning on using these to segregate
Hi Brian,
I've found having my spec test print out a catalog is a good way to build out a
suite of tests
context 'when the do_stuff param has the value of hell_yeah' do
let (:facts) {{'osfamily' = 'RainbowOS }}
let (:params) {{'do_stuff' = 'hell_yeah'}}
It 'print a catalog for spec
indeed.
I'd strongly encourage y'all to move to 3.5+ there's a bunch of really
nifty stuff that's been added... not to mention it's incredibly more
performant.
Say Hi to Seb for me, and pass along my contact info.
W
On Wed, May 21, 2014 at 12:13 AM, William Leese william.le...@meltwater.com
Hey guys,
I wrote a post on how to profile your puppet catalog compilations (3.4+) to
find resource hogs
Hope this helps someone
http://www.wolfspyre.com/2014/05/profiling-puppet-catalog-compilations/
--
You received this message because you are subscribed to the Google Groups
Puppet Users
at the very least, I'd alter the process you use to copy the new modules
dir into something like $puppetconfigdir/modules_MMDDYYSTUFF and then doing
2 mv's:
mv modulepathdir modulepathdir.old
mv modules_MMDDYYSTUFF modulepathdir
and then puppet should be good, you can start the rm -rf on module
I would LOVE it if this (or some other tool) would be capable of sending
puppetdb pushes to a back-of-house / reporting puppetdb instance
asynchronously... this seems like a nifty feature.. course, being able to
list multiple puppetdb instances in /etc/puppet/puppetdb.conf might be just
as nice
Hi All,
I am going to be doing a large bit of module refactoring, without an
existing suite of tests; and I wanted to convince a master inside of
vagrant on my dev laptop to compile a catalog against a collection of node
names to try to catch breaking changes more expediently and less impact
Hi Santanu,
custom facts should always do their work inside a setcode block.
http://docs.puppetlabs.com/guides/custom_facts.html
That has nothing to do with the issue at hand, however.
If you need the file to exist before puppet's first run, have your
provisioning system lay it down. Otherwise,
I'd be interested in accomplishing a similar thing. I want to have a
central reporting puppetdb which I can (outside of puppet) query for
multisite information without tying isolated environments together via a
single postgres db, or using multi-master replication.
my needs do not involve making
Hi Paul,
Here's a diagram showing how the puppet run process flows:
http://www.aosabook.org/images/puppet/TimingDiagram.png
as you can see, facter is run exactly once, before the catalog is created.
facter is not invoked again until the next run. I suppose you could have
your sshkey resource
Hi Sergey,
I've noticed behavior similar to then when resource ordering is specified
outside of a class; something like this:
#foo class
#
#
Class['prereq'] - Class['foo']
class foo() {
#some stuff here
}
since that dependency isn't occurring INSIDE the boundary of Class['foo'],
there can be
Hi Sebastian,
A potentially better way to do it might be to set purge = true on the
/etc/yum.repos.d directory, and specifically manage every file/repo in
there via puppet. This should insure that there are no repos present that
puppet didn't create
HTH
w
On Thu, Sep 26, 2013 at 1:28 PM,
On Thu, Sep 26, 2013 at 5:21 PM, jcbollinger john.bollin...@stjude.orgwrote:
On Thursday, September 26, 2013 3:11:24 PM UTC-5, Wolf Noble wrote:
Hi Sergey,
I've noticed behavior similar to then when resource ordering is specified
outside of a class; something like this:
#foo class
When I ran into this, what bit me was funky characters hiding in comments
in the manifests.
On Tue, Sep 17, 2013 at 3:42 PM, Robin Lee Powell
rlpow...@digitalkingdom.org wrote:
As of upgrading master and clients to 3.3.0, some of my servers are
saying:
Error: Could not prefetch package
Glad to hear you got it worked out!
You might have good luck with this module for controlling network interfaces
https://github.com/adrienthebo/puppet-network
=MobileMail=
Please excuse brevity or spelling errors
On Sep 15, 2013, at 22:31, Mani Devarajan
Hi Sam,
it somewhat depends on your requirements, but if you have a window during
which changes are not permitted, end of story, what about a cronjob to set
it to false at a particular time?
or in the same run that you set noop to true, require an exec which creates
an at job to re-enable at a
It seems like the match is 'Scientific' not 'Scientific Linux' but I'm nowhere
near a computer to check
MobileMail
On Jul 10, 2013, at 19:36, Stefan Lasiewski lasiew...@gmail.com wrote:
I am using Scientific Linux 6.4. Scientific Linux is a derivative of Red Hat
Enterprise Linux, similar
Hi Daniel,
What I've done in the past is:
Foo.erb
% foobar = scope.lookupvar('::somefact')
othervar = scope.lookupvar('class::variable')
lastvar = scope.lookupvar('myclass::myvar')
fqdn = scope.lookupvar('::fqdn')
-%
#this file generated automatically by puppet for %=fqdn%
#manual changes
Hi Stefan,
I've run into similar issues before… I don't have module code handy, but
what I ended up doing was creating an exec which performed something
similar to:
exec{'nuke_pesky_pkg':
command = 'rpm -e $package; touch /etc/package.removed',
creates = '/etc/package.removed'
}
which I
Hi Stack,
here's an example:
https://github.com/wolfspyre/puppet-mysqlx/blob/master/manifests/v55.pp
in this class I create a pair of bookend anchors, and then shove a yumrepo,
and a file resource between them.
Your other classes/resources can then declare a relationship to the end
anchor to
Hi Amos,
Here's what I've done:
in spec/spec_helper.rb:
Puppet.settings[:confdir] = spec/fixtures
then in the class I was working on, I made a parameter:
class foo(
$puppetdir = $settings::confdir
){
…
}
then in that class's spec test:
require 'spec_helper'
require
Hey Gang,
I've been asked to explore the landscape of network device management via
puppet.
in particular, I'm looking to support cisco, and force10 switches.
I know about the network device system, I'm wondering if that's the way to
go for both cisco and force10, or if there's a better way for
snip
It is incorrect to make that relationship declaration (via the chain
operator) if there are any nodes that do not declare the specified anchor,
because top-level declarations in *any* manifest file apply to all
nodes. It is poor practice to make declarations at top level like that
other
I'd like to say I really like the intent behind this.
Yes, a name is important. So is continuity. I believe the underlying goal is to
provide a collection of reference material. Once the skeleton of this material
is generally agreed upon, it can be used as guidelines to the community for how
Hi JJ,
I've seen this happen when class dependencies are placed outside the
specific class' scope, ie:
# default.pp
#
#
Anchor['my_prereq::end'] - Class['default']
class{ 'default':
notice 'default'
}
I've only seen this occur with anchors, but it's entirely possible it
happens with other
Hi Dan,
I've a script here that grabs the mysql 5.5 / 5.6 packages and turns them
into a yumrepo for easier consumption:
https://github.com/wolfspyre/MySQL_Createrepo
it's not perfect, but imho better than grabbing the RPMs manually form
Oracle.
On Wed, May 15, 2013 at 3:24 PM, Dan White
Hi Matt,
What I've seen done before is to use ${modulename}x for 'wrapper' modules
that give you the ability to stay current with the
forge/release/externally-owned module, but still maintain the necessary
site-spectific-shim code to extend the functionality of the original module
in the ways you
It seems to me that you will be relinquishing puppet's ability to maintain the
content of the file by doing this. If that's acceptable, use the replace
parameter to only have puppet maintain the initial config and allow the file to
be changed outside puppet moving forward. If that's not
Hi Dominic,
take a look at the vcsrepo module:
http://forge.puppetlabs.com/puppetlabs/vcsrepo
this should likely get you everything you need.
On Mon, May 20, 2013 at 2:45 PM, DJames dominicrja...@gmail.com wrote:
How can i setup puppet to pull code from A current SVN server?
--
You
Hi Ken,
I'm happy to give my $.02, FWIW.
I've found immense benefit from the overall paradigm described in Craig
Dunn's blog post here:
http://www.craigdunn.org/2012/05/239/
That paradigm, combined with modules which have all of their variables
placed as module::variable_name parameters which
Hi Simon,
On Tue, Apr 30, 2013 at 12:08 PM, Simon Piette piette.si...@gmail.comwrote:
Hi Wolfe,
On Tue, Apr 30, 2013 at 12:05 PM, Wolf Noble wno...@datapipe.com wrote:
Hi Simon,
On Apr 29, 2013, at 7:45 PM, Simon Piette piette.si...@gmail.com
wrote:
What I've done elsewhere is to make
Hi Simon,
On Apr 29, 2013, at 7:45 PM, Simon Piette piette.si...@gmail.com
wrote:
Hi all,
snip
Any solution for testing variables that depends on settings will be welcome.
Thanks,
I sent this last night from another account, but it didn't seem to go through:
What I've done elsewhere is
Hi Simon,
On Apr 29, 2013, at 23:35, Simon Piette piette.si...@gmail.com wrote:
Hi all,
snip
Any solution for testing variables that depends on settings will be welcome
What I've done elsewhere is to make a parameter for, in your case, vardir:
(untested, but close)
Class foo($vardir =
On Apr 25, 2013, at 9:47 AM, jcbollinger john.bollin...@stjude.org
wrote:
On Wednesday, April 24, 2013 4:24:08 PM UTC-5, Larry Fast wrote:
I keep running into the following upgrade pattern.
Previous Puppet run declared:
file { my_jar.1.2.3.jar: ... }
New puppet run declares:
What we've been doing here is trying to make all our hiera calls as class
parameter default values. This allows us to set param values in the unit tests,
thus bypassing hiera.
This has required some refactoring, but it has the advantage of being a simpler
test collection to implement.
Does
Hi John,
Thanks for taking the time to respond :) Replies inline.
On Mar 20, 2013, at 8:01 AM, jcbollinger john.bollin...@stjude.org
wrote:
On Tuesday, March 19, 2013 2:27:22 PM UTC-5, Wolf Noble wrote:
Hi there,
in house, we rely on the lsb facts:
lsbdistcodename, lsbdistdescription
Hi Chris,
Take a look at run stages
http://docs.puppetlabs.com/guides/language_guide.html#stages
They make it a little easier to perform some coarse ordering.
additionally, you can use PuppetLabs stdlib module and take advantage of the
anchor pattern sparingly.
Here's a good example of its use:
Hi there,
in house, we rely on the lsb facts:
lsbdistcodename, lsbdistdescription, lsbdistid, lsbdistrelease, and
lsbmajdistrelease
all of which are essentially derived from the
/etc/${distro-specific-release-file} and don't _really_ need the lsb packages
(and the associated bloat) installed.
Did you try removing the cert from a node and seeing if that changes the
behavior? you removed the certs from the master, but the node still thinks it
has a valid cert maybe?
This message may contain confidential or privileged information. If you are not
the
On Jan 30, 2013, at 4:52 AM, Sylvain Mougenot sylvain.mouge...@gmail.com
wrote:
Hello, I'm new to puppet using a puppet-master. And I couldn't figure out
what's the best practice I need.
My context :
- puppet-master with a puppet dashboard is setup and running
- few modules are ready and
On Jan 30, 2013, at 2:56 PM, Jason Cochard jason.coch...@gmail.com
wrote:
Hello all,
We are in the process of moving our data center from one city to another city
(far away). This won't be a physical move -- we are simply bringing up
equivalent machines in the new location and then
Hi All,
I noticed that a few years back, someone from Puppet Labs (well Reductive Labs
at that time) reached out to the openscap list in an attempt to collaborate on
puppet modules:
https://www.redhat.com/archives/open-scap-list/2010-March/msg0.html
it seems like the aqueduct project
On Jan 22, 2013, at 3:04 PM, jcbollinger john.bollin...@stjude.org wrote:
On Tuesday, January 22, 2013 7:08:09 AM UTC-6, Boyan Tabakov wrote:
Hello,
Let's consider the scenario when a client node in a puppet environment
gets compromised.
In case some of the puppet modules make decisions
This is actually a module I would love to see CURATED by PL, as I believe it is
a great real-world example to demonstrate the current recommendations from
which many people can learn.
The focus is pretty tight, however it could make a lot of people's lives easier.
Many community
Hi Gang,
My colleagues and I are contemplating refactoring our modules to take advantage
of the roles/profiles paradigm suggested by Craig Dunn in his blog post found
here:
http://www.craigdunn.org/2012/05/239/
Before we jump feet-first into adopting this paradigm, I thought it a good idea
to
| -%
group %=hash['group']-% v2c %=key%
%end-%
# Filesystem monitoring enabled
% filesystems.each_pair do |key, hash| -%
# %=key%
disk %=hash['mountpoint']-% %=hash['threshold']%
%end-%
does that help at all?
Wolf Noble
UNIX Team Lead
Datapipe Managed IT Services
1.201.792.4847 (international
get too much further
Thanks so much in advance
Wolf Noble
This message may contain confidential or privileged information. If you are not
the intended recipient, please advise us immediately and delete this message.
See http://www.datapipe.com/legal
Hi Gary,
I know what you mean, and agree for modules destined to be released to the wild.
unfortunately, we have a diverse enough environment that even the defaults
occasionally need to be overridden; hence the hierification.
(I've got all that logic wrapped into a case statement based on if
Hello all,
Is anyone managing custom static routes via puppet? if so, how?
I'm wondering if there's a better cross-platform way of adding routes than a
custom init script that defines the routes that need to be associated with each
interface…
or maybe someone with extra tasty brains (zombies
Hi Keiran,
I wrote this and posted it to the list awhile ago:
https://github.com/wolfspyre/vmware_puppetfact
there's been some suggestion that the source I used as a basis for this fact is
slightly inaccurate, however I don't have a suite of esx servers to test on..
(it's on my list)
feel
Hi there!
Is there a timeline on when the puppetlabs-firewall forge module will be
updated (not necessarily to 1.0) ? the github version has the firewallchain
type (as well as other nifty) that the forge version seems to be lacking, but
it's not clear from my perusing that the two are
I've just added a fact to github which on vmware hosts running linux runs
dmidecode and matches known bios address/ esx version to realize an ESX version
fact.
this information generated from
http://virtwo.blogspot.com/2010/10/which-esx-version-am-i-running-on.html
in limited testing on an
I believe hiera_array() collects and provides an array of all the relevant
elements up the entirety of the tree. To collect only the most relevant data
just use hiera()
On Sep 3, 2012, at 10:21 AM, Josh j...@chickenmonkey.co.uk
wrote:
Versions: puppet 2.7.18, hiera 0.3.0
I have
why not check to see if the binary which you're using to talk to the raid hw
exists.. if so, execute and provide facts, if not, do nothing..
That's how it's being done in the hpacucli fact I've been using.
On Aug 23, 2012, at 11:26 AM, Nigel Kersten ni...@puppetlabs.com
wrote:
On Thu, Aug
I accomplished something similar awhile back.
While I've not spent much time looking at your particular problem, I think
this'll help point you in the direction of what you need…
class mymodule::params{
$collector= hiera('mymodule_collector','')
}
class mymodule::config{
$collector
I was asleep, and woke up thinking about a way to define a relationship between
the hiera parameters of multiple modules such that conflicts could be avoided...
The thought process was that if I set one value, say, the service enablement
parameter for snmp to 'stopped' in my tripwire module,
if you're using hiera, why not something like:
foo_data_webfarm_http_servers:
foo1.example.com: {
ip: '1.2.3.4',
port: '80',
}
foo2.example.com: {
ip: '2.3.4.5',
port: '80',
}
foo_data_webfarm_https_servers:
foo1.example.com: {
ip: '1.2.3.4',
port: '443',
}
On Aug 9, 2012, at 5:42 PM, Tim Mooney tim.moo...@ndsu.edu
wrote:
In regard to: Re: [Puppet Users] how to resolve hostnames to IP addresses...:
if you're using hiera, why not something like:
[…]
heh yeah,
okay, so, well, uh... I didn't say it was a GOOD idea ;)
Thanks for the
is that none of it has tried to contact us.”
Bill Waterson (Calvin Hobbes)
- Wolf Noble wno...@datapipe.com wrote:
This message may contain confidential or privileged information. If you are not
the intended recipient, please advise us immediately and delete
I wonder…
Would there be a benefit in the notion of 'peer values' in hiera?
Currently hiera_array will collect all the possible values for $thing defined
all the way up the hierarchy…
This js awesome.
But there's no way to find out the values of $thing for other nodes at the same
level in the
I'm doing something similar in one environment… I've come to realize this
method contains less awesome than I'd anticipated.
What I'd suggest doing, is changing the file deployed to the client into a
template that's deployed via puppet..
have that template set the content correctly for the
Hi gang,
I feel like I'm missing something fundamental here…
I've got the following:
class snmp::rhel::rh5enable {
$collector = ['10.0.0.1', '10.0.0.2']
define add_snmp_hosts_allow ($ip) {
exec { hosts_allow_$ip:
command = /bin/echo \snmpd : $ip : ALLOW\ /etc/hosts.allow,
but not using parameters.
-Eric
--
Eric Shamow
Professional Services
http://puppetlabs.com/
(c)631.871.6441
On Sunday, July 8, 2012 at 2:00 PM, Wolf Noble wrote:
Hi gang,
I feel like I'm missing something fundamental here…
I've got the following:
class snmp::rhel::rh5enable {
$collector
, at 2:38 PM, Wolf Noble wrote:
Hi Eric,
No, this class is not being called anywhere else.
What I think is going on is that I have an array of two strings, but somehow
the defined type is getting a concatenated string which contains both values
of the array instead of treating each element
This thought crossed our minds as well.
I created the following feature request awhile back:
https://projects.puppetlabs.com/issues/13934#change-60706
the intent behind which is if this fact is flagged as immutable, and it
changes, something is drastically wrong… (don't)? do something
until
Hi Jakov,
the two semicolons locate these variables as being in top-scope/variables
outside of any specific module.
otherwise it might be $modulename::variable
Does that help at all?
W
On Jun 18, 2012, at 8:25 AM, Jakov Sosic wrote:
Hi.
I have the following facts available:
# facter |
D'oh
On Jun 18, 2012, at 11:26 AM, Martin Alfke wrote:
Hi,
On 18.06.2012, at 17:14, Wolf Noble wrote:
Hi Jakov,
the two semicolons locate these variables as being in top-scope/variables
outside of any specific module.
please note that these are colons (:) no semicolons(;).
I
… Now that I'm excessively caffeinated (I'm not actually sure if that's better,
or worse, than my previous state, but)…
On Jun 18, 2012, at 12:47 PM, Jo Rhett wrote:
On Jun 18, 2012, at 12:55 AM, Felix Frank wrote:
this thread is starting to confuse me, I am no longer sure what you're
On Jun 18, 2012, at 6:04 PM, Jo Rhett wrote:
On Jun 18, 2012, at 1:57 PM, Wolf Noble wrote:
Well, a in the service of b -- but as a general point, I think that every
notify/subscribe should be tune-able as to which things changing will cause
the action to take place.
Not to continue
I had some issues with using boolean logic with data being fed to me by hiera.
I went for a slightly hacky option of:
might help you, might be a BadIdea(™)
(untested)
class foo::params{
bool_to_be_tempname = hiera('some_TorF_key_as_a_string')
case bool_to_be_tempname {
[tT][rR][uU][eE]:
where the pointy/sharp parts of this might be.
On Jun 13, 2012, at 8:24 AM, jcbollinger wrote:
On Tuesday, June 12, 2012 5:51:06 PM UTC-5, Wolf Noble wrote:
Hi John,
Thanks a bunch. This seems incredibly sensible.
If I'm understanding your suggestion properly, I interpret the best thing
So… I'm coming across a situation where I foresee myself having a collection of
related k/v pairs in hiera. I want to create a thing… lets say a yumrepo for
example's sake, but this same idea translates to a vhost, a web app
configuration , a virtual domain, etc.
lets assume a hierarchy of:
?
On Jun 12, 2012, at 5:18 PM, jcbollinger wrote:
On Tuesday, June 12, 2012 1:48:09 PM UTC-5, Wolf Noble wrote:
So… I'm coming across a situation where I foresee myself having a collection
of related k/v pairs in hiera. I want to create a thing… lets say a yumrepo
for example's sake
so I've set a hiera key, and trying to display some different data in motd
depending on if the value of said key is true or false… it gets set
accordingly.. but my motd template doesn't seem to be properly reacting when
the value of it changes..
it's as if the variable never matches true
not
Thanks Gary,
I knew it was something simple.
Appreciate the help
% if foostatus == true -%
This is the problem - you're trying to compare a boolean value of true to a
string value of 'true'. Just do something like:
% if foostatus -%
That should work for you :)
Hi Luke,
Would adding fqdn to the bottom of your hierarchy give you the ability to
override the values on the outliers that you can't automagically set while
retaining elegance in your class? Or am I missing something?
Wolf Noble
Datapipe Managed IT Services
On May 11, 2012, at 12:45 PM
/puppet-users?hl=en.
Wolf Noble
Senior UNIX Systems administrator
Datapipe Managed IT Services
Tel: 201.792.1918 x2910
This message may contain confidential or privileged information. If you are not
the intended recipient, please advise us immediately
Hi Gang,
If you're a mac user and haven't heard of text bucket, but find yourself typing
the same things semi-often.. you might want to check it out.
http://bassetsoftware.com/b.cgi/textbucket/index
I started using it about 9 months ago when someone showed me klypper on linux
and I went
I'd enjoy coming to this. Please keep me posted as things develop.
Thanks Dan!
MobileMail
On Feb 7, 2012, at 7:05 PM, Dan Bode
d...@puppetlabs.commailto:d...@puppetlabs.com wrote:
Hi Austin Puppet Users,
I will be in the area in a few weeks and I would like to try to get some of the
local
just a thought; what if you made a fact for number of apache workers currently
serving something, and base the service restart off of that.
Might not be the best option; might not even work, but it makes more sense to
me to have the service restart dependent on a local fact than be dependent on
Hi Peter,
we used a different method here for linux hosts.
We put the groups we want to grant access to in /etc/security/access.conf ; ala:
[root@---]# egrep -v ^# /etc/security/access.conf
- : ALL EXCEPT root admin pci_sysadmin pci_devadmin : ALL
and then just add users to the group which
Here's what I've been doing, fwiw:
file { 'nsswitch.conf':
path = '/etc/nsswitch.conf',
content = inline_template(
file(
/etc/puppet/modules/dist/ldapclient/templates/etc/nsswitch.conf.$hostname.erb,
84 matches
Mail list logo
