Re: [Puppet Users] yum.puppetlabs.com rsync problem
rsync to yum.puppetlabs.com didn`t work now =( 22/tcp open ssh 25/tcp closed smtp 80/tcp open http 443/tcp open https 631/tcp closed ipp /tcp closed unknown 5666/tcp open nrpe вторник, 19 июля 2011 г., 10:54:48 UTC+4 пользователь MoonWolf написал: Zach Leslie wrote: The rsync config should be sorted now. Give a test and let us know if there are any issues. It works like a charm now, thank you! -- Best regards Marcin Trendota -- You received this message because you are subscribed to the Google Groups Puppet Users group. To view this discussion on the web visit https://groups.google.com/d/msg/puppet-users/-/dLlwwzBB2VAJ. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
[Puppet Users] ERB template, both client and master data
Hi all, I'm trying to build a motd template for all my linux machines. This should contain two type of variables: - client side (from facter), like operatingsystem and architecture - supplied by the master, after parsing a text file (or querying a mysql db), like description and location. So. Where are the template variables evaluated? On the client or on the master? Any hints on how I can do this? Thanks. -- You received this message because you are subscribed to the Google Groups Puppet Users group. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
Re: [Puppet Users] Re: Puppet 3.0.0 and Hiera
- Original Message - From: jcbollinger john.bollin...@stjude.org To: puppet-users@googlegroups.com Sent: Monday, October 15, 2012 9:45:12 PM Subject: Re: [Puppet Users] Re: Puppet 3.0.0 and Hiera snip Indeed, I personally look at it quite the other way around: hiera integration makes parametrized classes tolerable. It provides a good I agree, tolerable is the right word here, was not suggesting the integration now makes them totally usable. however with a simple consistent approach and style in your own modules you can work around some of the remaining issues. We knew dynamic scoping sucked yet most of us have come up with a way to write our modules so we don't get bit by those issues by adopting a few simple rules and avoid some of the more problematic areas. Applying the same approach to param classes did at least get me to a place that I feel was better off than the older method - but yes, its a shame we even have to come up with a style for working around problems in something as key as class paramaterization. -- You received this message because you are subscribed to the Google Groups Puppet Users group. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
[Puppet Users] Re: ERB template, both client and master data
The template function is evaluated on the server, but you do have access to the facter variables that are generated on the client. Kind regards, Ger. Op dinsdag 16 oktober 2012 09:53:34 UTC+2 schreef SAF het volgende: Hi all, I'm trying to build a motd template for all my linux machines. This should contain two type of variables: - client side (from facter), like operatingsystem and architecture - supplied by the master, after parsing a text file (or querying a mysql db), like description and location. So. Where are the template variables evaluated? On the client or on the master? Any hints on how I can do this? Thanks. -- You received this message because you are subscribed to the Google Groups Puppet Users group. To view this discussion on the web visit https://groups.google.com/d/msg/puppet-users/-/4K0VmNECkSoJ. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
Re: [Puppet Users] augeas only add if doesn't exist
On 15/10/12 16:52, Eugene Vilensky wrote:
On Oct 14, 2012, at 12:40 PM, Dominic Cleal dcl...@redhat.com
mailto:dcl...@redhat.com wrote:
You're missing quotes around nrpe, the command should be:
set /files/etc/hosts.allow/*[process='nrpe']/client[last()+1] 1.1.1.1
rodjek's example has the quotes in, so just a transcription error.
Totally works for an existing 'nrpe' node (thanks!) but not if it does
not exist. I don't suppose there is a simple create onlyif not exists
expression? (http://augeas.net/page/Path_expressions)
The best way to do this currently is two separate resources, splitting
out the responsibilities so that one adds nrpe and the client
1.1.1.1 if the process isn't there already. The second, if you need
it, will run if nrpe is already listed but the client 1.1.1.1 isn't.
$process = nrpe
$client = 1.1.1.1
# Responsible for adding nrpe if it isn't there
augeas { process-${process}:
context = /files/etc/hosts.allow,
changes = [
set /files/etc/hosts.allow/01/process ${process},
set /files/etc/hosts.allow/01/client[.='${client}'] ${client},
],
onlyif = match *[process='${process}'] size == 0,
}
# Responsible for updating existing nrpe entries missing the client
augeas { process-${process}-client:
context = /files/etc/hosts.allow,
changes = set
/files/etc/hosts.allow/*[process='${process}']/client[.='${client}']
${client},
require = Augeas[process-${process}],
}
I've changed this from last()+1 to a style that makes the command
idempotent.
--
Dominic Cleal
Red Hat Consulting
m: +44 (0)7817 878113
--
You received this message because you are subscribed to the Google Groups
Puppet Users group.
To post to this group, send email to puppet-users@googlegroups.com.
To unsubscribe from this group, send email to
puppet-users+unsubscr...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/puppet-users?hl=en.
Re: [Puppet Users] augeas only add if doesn't exist
On 16/10/12 11:42, Dominic Cleal wrote:
On 15/10/12 16:52, Eugene Vilensky wrote:
On Oct 14, 2012, at 12:40 PM, Dominic Cleal dcl...@redhat.com
mailto:dcl...@redhat.com wrote:
You're missing quotes around nrpe, the command should be:
set /files/etc/hosts.allow/*[process='nrpe']/client[last()+1] 1.1.1.1
rodjek's example has the quotes in, so just a transcription error.
Totally works for an existing 'nrpe' node (thanks!) but not if it does
not exist. I don't suppose there is a simple create onlyif not exists
expression? (http://augeas.net/page/Path_expressions)
The best way to do this currently is two separate resources, splitting
out the responsibilities so that one adds nrpe and the client
1.1.1.1 if the process isn't there already. The second, if you need
it, will run if nrpe is already listed but the client 1.1.1.1 isn't.
Sorry, that wasn't a great example. I'd set context but not used it.
$process = nrpe
$client = 1.1.1.1
# Responsible for adding nrpe if it isn't there
augeas { process-${process}:
context = /files/etc/hosts.allow,
changes = [
set 01/process ${process},
set 01/client[.='${client}'] ${client},
],
onlyif = match *[process='${process}'] size == 0,
}
# Responsible for updating existing nrpe entries missing the client
augeas { process-${process}-client:
context = /files/etc/hosts.allow,
changes = set *[process='${process}']/client[.='${client}'] ${client},
require = Augeas[process-${process}],
}
--
Dominic Cleal
Red Hat Consulting
m: +44 (0)7817 878113
--
You received this message because you are subscribed to the Google Groups
Puppet Users group.
To post to this group, send email to puppet-users@googlegroups.com.
To unsubscribe from this group, send email to
puppet-users+unsubscr...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/puppet-users?hl=en.
[Puppet Users] Re: $concat_basedir not defined
hi. i fixed it after read the docs:
Enabling Pluginsync
After setting up the directory structure, we then need to turn on
pluginsync in our puppet.conf configuration file on *both the master and
the clients:*
[main]
pluginsync = true
On Tuesday, July 31, 2012 11:06:04 PM UTC+7, Shannon McFarland wrote:
Hey folks,
I am using puppet for some OpenStack deployments and on a new node things
look pretty good right until after the successful reception of a cert from
the puppet master. I get this error:
err: Could not retrieve catalog from remote server: Error 400 on SERVER:
$concat_basedir not defined. Try running again with pluginsync enabled at
/usr/share/puppet/modules/concat/manifests/setup.pp:25 on node
controller-1.example.com
warning: Not using cache on failed catalog
err: Could not retrieve catalog; skipping run
Things seem to be as they should in the setup.pp file it refers to. I
have searched for similar issues and cannot seem to find much on this
issue. It happens every time I build a node.
Below is the setup.pp file for your reference.
# Sets up the concat system.
#
# $concatdir is where the fragments live and is set on the fact
concat_basedir.
# Since puppet should always manage files in $concatdir and they should
# not be deleted ever, /tmp is not an option.
#
# $puppetversion should be either 24 or 25 to enable a 24 compatible
# mode, in 24 mode you might see phantom notifies this is a side effect
# of the method we use to clear the fragments directory.
#
# The regular expression below will try to figure out your puppet version
# but this code will only work in 0.24.8 and newer.
#
# It also copies out the concatfragments.sh file to ${concatdir}/bin
class concat::setup {
$id = $::id
$root_group = $id ? {
root= 0,
default = $id
}
if $::concat_basedir {
$concatdir = $::concat_basedir
} else {
fail (\$concat_basedir not defined. Try running again with pluginsync
enabled)
}
$majorversion = regsubst($::puppetversion,
'^[0-9]+[.]([0-9]+)[.][0-9]+$', '\1')
file{${concatdir}/bin/concatfragments.sh:
owner = $id,
group = $root_group,
mode = '0755',
source = $majorversion ? {
24 = 'puppet:///concat/concatfragments.sh',
default = 'puppet:///modules/concat/concatfragments.sh'
};
[ $concatdir, ${concatdir}/bin ]:
ensure = directory,
owner = $id,
group = $root_group,
mode = '0750';
## Old versions of this module used a different path.
'/usr/local/bin/concatfragments.sh':
ensure = absent;
}
}
--
You received this message because you are subscribed to the Google Groups
Puppet Users group.
To view this discussion on the web visit
https://groups.google.com/d/msg/puppet-users/-/ZL-kqVBZNCYJ.
To post to this group, send email to puppet-users@googlegroups.com.
To unsubscribe from this group, send email to
puppet-users+unsubscr...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/puppet-users?hl=en.
[Puppet Users] Setting up Java on Windows
Hi guys,
I have been trying to bootstrap a windows machine with puppet. I need to
install java and setup the JAVA_HOME and add bin to the path variable of
windows.
I have tried using the package resource with the jdk setup but everytime I
run the manifest puppet opens the setup wizard and waits for me to do
something.
Am I doing something wrong or is this the right behavior? Also is there a
way to do this as silent install?
Here is my manifest.
package {java :
ensure = installed,
provider = windows,
source = C:/Manifests/installables/jdk-6u35-windows-i586.exe,
}
Thank you.
--
You received this message because you are subscribed to the Google Groups
Puppet Users group.
To view this discussion on the web visit
https://groups.google.com/d/msg/puppet-users/-/h75UozzbWpAJ.
To post to this group, send email to puppet-users@googlegroups.com.
To unsubscribe from this group, send email to
puppet-users+unsubscr...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/puppet-users?hl=en.
[Puppet Users] Re: The free software tarballs are now difficult to find
On Sunday, October 14, 2012 7:26:23 PM UTC-5, Nigel Kersten wrote: On Saturday, October 13, 2012 8:23:57 AM UTC-7, Paul Belanger wrote: On Friday, October 12, 2012 10:10:54 AM UTC-4, windowsrefund wrote: Recently, there have been some changes made to the Puppetlabs website which result in the free software releases being difficult to locate and download. Visitors using the download links are taken directly to the non-free Enterprise option rather than being presented with an option. Personally, I'd like to see the site reverted so users have the option. FYI, the wiki section still provides this useful content here: http://projects.puppetlabs.com/projects/1/wiki/downloading_puppet Best, Adam Kosmin I was asking this on IRC the other day. I understand the need to collect information for prospective customers, however requiring a pay wall (personal information) to access the open source tarballs does not seem to be in the spirit of free software[1]. There are lots of ways people get open source Puppet, and not many people actually grab them in tarball form from the website from the stats we have. Apart from being in the distributions, most of you seem to be getting software from the apt/yum repos, which don't require a paywall: http://docs.puppetlabs.com/guides/puppetlabs_package_repositories.html If you're hitting the main page, yes, the Download link goes to PE by default, with a link to the OSS download page at the bottom. The Products - Puppet Open Source link on the main page takes you here: http://puppetlabs.com/puppet/puppet-open-source/ where we promote getting the source from GitHub rather than tarballs directly from us. The reasoning has generally been that if you want to use Puppet, the best way of consuming it is via packages, and if you want to get the source, the best way of consuming it is via Git. That way updates are simpler for users, and development from source is simpler for developers. For what it's worth, I always prefer to get source for third-party software, and my absolute *least* favorite way to do so is from a source-control repository. I particularly like source RPMs, but tarballs generally work just fine for me. More generally, I want *one file*, associated with a specific release of the software and so labeled, and accessible via a general purpose network protocol such as HTTP or FTP from a location sponsored by the project. I don't see any dark conspiracy behind PL's website changes, but I am a bit surprised by some of the responses from PL personnel. The tone, if not the actual content, seems in some cases to say there's nothing wrong, why are you bothering us with this? Clearly something *is* wrong as far as Adam is concerned. PL is certainly not obligated to cater to any individual user's preferences -- or even to the whole community's preferences -- but it is not helpful to anyone to try, as some of the responses seemed to do, to deny their validity. Given a complaint that the source tarballs are hard to find, I would have expected a response more along the lines of We're sorry the changes to our web site have inconvenienced you. We assure you that our intent is not to hide or block access to open-source Puppet. We will consider whether the site can be changed further to better serve both PE and our open-source Puppet users. I might even have expected a suggestion to file a ticket or a solicitation for further community comment. John -- You received this message because you are subscribed to the Google Groups Puppet Users group. To view this discussion on the web visit https://groups.google.com/d/msg/puppet-users/-/V7QOTlFLbZUJ. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
Re: [Puppet Users] Setting up Java on Windows
On Tue, Oct 16, 2012 at 5:45 AM, wukong wicked.wuk...@gmail.com wrote:
Also is there a way to do this as silent install?
Here is my manifest.
package {java :
ensure = installed,
provider = windows,
source = C:/Manifests/installables/jdk-6u35-windows-i586.exe,
}
Launching that .exe will always give you a GUI, unless you provide
some additional arguments. To install the JRE silently, see the
following:
http://www.oracle.com/technetwork/java/javase/silent-136552.html#running.
Regards,
Matt.
--
You received this message because you are subscribed to the Google Groups
Puppet Users group.
To post to this group, send email to puppet-users@googlegroups.com.
To unsubscribe from this group, send email to
puppet-users+unsubscr...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/puppet-users?hl=en.
[Puppet Users] Unable to create the rrd graph
Hello Everybody, I have been trying to use the rrdgraph reporting feature available in puppet. After performing all the instruction maintained on the site http://docs.puppetlabs.com/references/latest/report.html , the rrd directory for the agents get created but they are empty. When i check the log files in /var/log/messages, I get to see the following error Failed to set owner to '0': Operation not permitted - /var/lib/puppet/rrd/ftldwshost174.wsdev.citrix.com RRD library is missing; cannot store metrics I have installed the RRD lib and tools using the following commands :- yum install rrdtool-ruby I tried installing from the source from the following url:http://rubyforge.org/projects/rubyrrdtool/ But it tends to give an error in make Can anybody tell me how to resolve this issue. Thanks, Nishant -- You received this message because you are subscribed to the Google Groups Puppet Users group. To view this discussion on the web visit https://groups.google.com/d/msg/puppet-users/-/OTfEiPc3ANgJ. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
[Puppet Users] Re: Puppet 3.0.0 and Hiera
On Monday, October 15, 2012 4:32:28 PM UTC-5, Ellison Marks wrote: I don't intend to use parametrized declarations(Right now I'm simply using hiera_include in the default node for everything and loving it). I do however, very much like the convenience of the new syntax. In addition, my use case for default values in the puppet backend that shipped with the class wasn't terribly complex, so simply including a default value in the class itself serves my needs well. In my case, the new features are simply a more convenient way of doing what i was already doing(sans the array and hash functionality). I don't see it, but that's a question of style and personal preference. Other than the effort required to get from here to there, there is nothing inherently negative about your plan as you describe it. Please don't take my comments as personal criticism. I truly am interested in whether there are advantages to class parametrization that may not be evident to me, but I used your comments mostly as a springboard for the discussion that now ensues. Regulars around here know that I am a longtime critic of Puppet's parametrized class design. It is better in Puppet 3 -- a lot better, in fact -- but there are still some significant issues. I consider it a community service to publicize those issues, both to help people avoid tripping over them and to keep PL's feet to the fire. John -- You received this message because you are subscribed to the Google Groups Puppet Users group. To view this discussion on the web visit https://groups.google.com/d/msg/puppet-users/-/3YOfOaxcwUkJ. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
[Puppet Users] Re: The free software tarballs are now difficult to find
Business logic aside the biggest issue for me is that it is difficult to find the free software version without knowing any better. As mentioned in this thread most distributions handle packaging Puppet nicely, as does rubygems. However a very good example is puppet 3.0 rc. I had been following the release notes for 3.0 for quite some time and was excited that rc7 was released and rc8 were released around the time I attended PuppetConf. It's not so much the fact that the tarball itself is not accessible, it's that it's difficult to find. The download page has mention of the open source version at the way bottom of the page in minor print compared to the rest of the page. I feel we'd be far better represented with a page that links to PE / OS version equally. I'd even be happy if you wanted to foresake the tarballs, but at least make the availability of the open source version more obvious. (Once clicked bring to a page that directs us to github for all I care.) I've referred friends to check out Puppet and most of them say Oh, I downloaded the trial to test it! without realizing there is an open source version at the bottom of the page which is more than adequate (and in my opinion more efficient to learn on) for their testing needs. Ultimately I can agree it's becoming a bit cumbersome to find, but most importantly it's becoming more hidden from the average new user to the site. Again I understand from a business logic, but I believe you'd be served well enough by equally weighting the links on the download page. You'd keep both the enterpriser's happy and the zealots happy. (Relatively. :) -- You received this message because you are subscribed to the Google Groups Puppet Users group. To view this discussion on the web visit https://groups.google.com/d/msg/puppet-users/-/5PwwqDLs9dsJ. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
Re: [Puppet Users] Re: Puppet 3.0.0 and Hiera
On Monday, October 15, 2012 6:59:45 PM UTC-5, Nick Fagerlund wrote: On Monday, October 15, 2012 1:45:12 PM UTC-7, jcbollinger wrote: Indeed, I have taken a second look, and a third, and maybe more. I love the hiera integration with parametrized classes. It was a fabulous idea, as it makes it reasonable and safe to use existing parametrized classes (provided you use only 'include' or 'require' to declare them). Even with that, however, parametrized classes offer very little of value over non-parametrized analogs that implement the same hiera-based external data access. One could argue, perhaps, that there is a documentary advantage in parametrization, but I think that's poor justification for introducing functional (so to speak) elements to any class. That's what comments are for. But parameters expose that info to more than humans -- for example, you can use the resource_type REST endpoint today to get a list of all classes and their parameters and defaults. Not a lot is using that today, but I expect more and more things to start doing auto-discovery of parameters, since it's a really machine-friendly way to find out what a class wants. That's an interesting angle of which I was not aware. I'm not sure I see a use case beyond automatic documentation generation, however, and there are alternative approaches that could serve that particular purpose. Do you envision other specific uses? Also, have you considered that any explicit hiera lookups in the class body also represent machine-discoverable things that a class wants? So long as there is no plan to deprecate the hiera functions, any approach to identifying classes' data dependencies that does not account for explicit lookups is incomplete. John -- You received this message because you are subscribed to the Google Groups Puppet Users group. To view this discussion on the web visit https://groups.google.com/d/msg/puppet-users/-/u5u9iR5s_8MJ. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
Re: [Puppet Users] Re: Puppet 3.0.0 and Hiera
- Original Message - From: jcbollinger john.bollin...@stjude.org To: puppet-users@googlegroups.com Sent: Tuesday, October 16, 2012 3:11:55 PM Subject: Re: [Puppet Users] Re: Puppet 3.0.0 and Hiera snip But parameters expose that info to more than humans -- for example, you can use the resource_type REST endpoint today to get a list of all classes and their parameters and defaults. Not a lot is using that today, but I expect more and more things to start doing auto-discovery of parameters, since it's a really machine-friendly way to find out what a class wants. That's an interesting angle of which I was not aware. I'm not sure I see a use case beyond automatic documentation generation, however, and there are alternative approaches that could serve that particular purpose. Do you envision other specific uses? web interfaces that prompts for the right info when someone adds a class to a machine. It would be great if you can set your email client to send plain text emails as per list guidelines. -- You received this message because you are subscribed to the Google Groups Puppet Users group. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
[Puppet Users] Re: The free software tarballs are now difficult to find
On Oct 16, 6:12 am, jcbollinger john.bollin...@stjude.org wrote: For what it's worth, I always prefer to get source for third-party software, and my absolute *least* favorite way to do so is from a source-control repository. I particularly like source RPMs, but tarballs generally work just fine for me. More generally, I want *one file*, associated with a specific release of the software and so labeled, and accessible via a general purpose network protocol such as HTTP or FTP from a location sponsored by the project. It is available from GitHub, tagged for each release, for example: https://github.com/puppetlabs/puppet/zipball/3.0.0 It's also available via Source RPM from our repos. I don't see any dark conspiracy behind PL's website changes, but I am a bit surprised by some of the responses from PL personnel. The tone, if not the actual content, seems in some cases to say there's nothing wrong, why are you bothering us with this? Clearly something *is* wrong as far as Adam is concerned. PL is certainly not obligated to cater to any individual user's preferences -- or even to the whole community's preferences -- but it is not helpful to anyone to try, as some of the responses seemed to do, to deny their validity. I'm not denying their validity of the issue. I am questioning what exactly the issue being raised is and the extent that it's a key issue for the community. Given a complaint that the source tarballs are hard to find, I would have expected a response more along the lines of We're sorry the changes to our web site have inconvenienced you. We assure you that our intent is not to hide or block access to open-source Puppet. We will consider whether the site can be changed further to better serve both PE and our open-source Puppet users. I might even have expected a suggestion to file a ticket or a solicitation for further community comment. My analysis of the situation was that this was less about the location of tarballs or access to software (given the numerous other ways you can get the software) but rather about the definition of free. Here some members of the community and Puppet Labs clearly differ. I believe, no matter what changes we make to our website, that the only way they will be satisfied is we return to a GPL-esque license. A change we're not going to make for a number of legitimate, documented, transparent business reasons. That limits our ability to have a productive conversation about these issues. That being said we are discussing internally whether to continue with this approach. We've not yet made a decision but I will communicate that decision either way when it's made. Kind Regards James Turnbull -- You received this message because you are subscribed to the Google Groups Puppet Users group. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
Re: [Puppet Users] Re: Puppet 3.0.0 and Hiera
On Tuesday, October 16, 2012 3:23:14 AM UTC-5, R.I. Pienaar wrote: its a shame we even have to come up with a style for working around problems in something as key as class paramaterization. I think it's a shame that prevailing opinion holds class parametrization to be a key feature. There continue to be problems with it largely because it is conceptually inconsistent with Puppet's model for classes (specifically, with their singleton nature). Because Puppet cannot accept inconsistent declarations of a class, Puppet DSL should not facilitate users writing such declarations into their manifests. Indeed, isn't that much the same logic PL used in deciding to remove dynamic scoping? Here's one solution: keep parametrized classes themselves, but deprecate and eventually remove the parametrized-class declaration syntax. That would leave class parameters as formalized declarations of external data used by classes, while removing the possibility of inconsistent class declarations. It would also pull back from the confusing effort to make classes appear to be resources when in fact they are not (for example, create_resources() won't create classes). John -- You received this message because you are subscribed to the Google Groups Puppet Users group. To view this discussion on the web visit https://groups.google.com/d/msg/puppet-users/-/_jiSN1JwQvMJ. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
Re: [Puppet Users] Re: Puppet 3.0.0 and Hiera
On Tuesday, October 16, 2012 9:16:43 AM UTC-5, R.I. Pienaar wrote: It would be great if you can set your email client to send plain text emails as per list guidelines. If I were using an e-mail client then I would be happy to do so, but I'm posting via Google's web interface. John -- You received this message because you are subscribed to the Google Groups Puppet Users group. To view this discussion on the web visit https://groups.google.com/d/msg/puppet-users/-/eNDqDeSzLn8J. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
[Puppet Users] PHP modules
Good day. I would like to ask. Does anyone have an elegant way to manage the PHP modules that are not available via a repository, and that can only be installed via pecl or pear. Ideally, it would actually be better if Debians dh-make-(pear|pecl) was actually used in the manifest. Github is not is not able to help. Kind Regards Brent Clark -- You received this message because you are subscribed to the Google Groups Puppet Users group. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
[Puppet Users] Re: Puppet 3.0.0 and Hiera
No worries. I'm enjoying the discussion. On Tuesday, October 16, 2012 7:01:47 AM UTC-7, jcbollinger wrote: On Monday, October 15, 2012 4:32:28 PM UTC-5, Ellison Marks wrote: I don't intend to use parametrized declarations(Right now I'm simply using hiera_include in the default node for everything and loving it). I do however, very much like the convenience of the new syntax. In addition, my use case for default values in the puppet backend that shipped with the class wasn't terribly complex, so simply including a default value in the class itself serves my needs well. In my case, the new features are simply a more convenient way of doing what i was already doing(sans the array and hash functionality). I don't see it, but that's a question of style and personal preference. Other than the effort required to get from here to there, there is nothing inherently negative about your plan as you describe it. Please don't take my comments as personal criticism. I truly am interested in whether there are advantages to class parametrization that may not be evident to me, but I used your comments mostly as a springboard for the discussion that now ensues. Regulars around here know that I am a longtime critic of Puppet's parametrized class design. It is better in Puppet 3 -- a lot better, in fact -- but there are still some significant issues. I consider it a community service to publicize those issues, both to help people avoid tripping over them and to keep PL's feet to the fire. John -- You received this message because you are subscribed to the Google Groups Puppet Users group. To view this discussion on the web visit https://groups.google.com/d/msg/puppet-users/-/e23mEImUQ0oJ. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
Re: [Puppet Users] Re: Wrapper classes, ordering anchors
On Thursday, October 11, 2012 6:27:07 PM UTC-5, Jeff McCune wrote:
Yeah, but sometimes the need to manage both sides of the wrapper class is
un-avoidable. I often find this to be the case when I'm dealing with
interpreter packages like java, ruby, perl, python, etc... The classes
that setup the packaging repositories need to happen before the class
managing the ruby / java / python / go / whatever interpreter... and the
classes that actually manage your application need to come after the class
that manages the interpreter.
So the interpreters are what is modeled in the wrapper class in this
discussion.
This pattern is necessary in other situations too; web servers that host
applications for example. Basically, I think to myself, I may need to
anchor down this class... if the thing the class is modeling sits between
the infrastructure layer and the application layer.
So, not quite as common as a bash script, but not so uncommon that you
never need to use it.
Again, the fact that it's necessary at all is definitely a bug in Puppet.
We just haven't fixed it yet, but not for lack of people trying. =(
It struck me just now that the conversation never touched on the 'include'
function's neglected little brother, 'require'. The 'require' function
does everything 'include' does, plus sets up a relationship that the
required class be applied before the requiring one. That's half of the
effect of the anchor pattern right there.
One could imagine a complement to the 'require' function that was the same
except for setting up an oppositely-directed relationship. For continued
parallelism with resource metaparameters, one might name such a function
'before'. I'm not convinced that such a function would have even as much
use as little-used 'require', but at least there would be symmetry.
The same theme does not work for full containment of one class by another,
because classes appear as single nodes in the relationship graph. You
would need separate nodes for the beginning of a class and its end to model
containment directly in the graph. I have heard talk before about
implementing a true container formalism for Puppet, but inasmuch as such a
thing has not yet appeared, perhaps an alternative would be a better target.
Consider, then, another potential new function, contain. Like 'require',
it would have the all the effects of include, but it would have the
additional effect of copying all relationships to or from the containing
class to corresponding relationships with the contained class -- much the
same way that relationships affect classes' resources.
That would not require any new DSL syntax or object type, nor any change to
the graph analysis code. I think it would support multiple containment for
no additional (development) cost. It would allow users easily to control
whether classes are contained or merely associated, and I think it could be
applied even when classes are declared via (evil) parametrized-class
syntax. That is, inasmuch as this works:
class { 'foo': }
include 'foo'
there's no reason why this would not work too:
class { 'foo': }
contain 'foo'
The only down side I see right now is an increase in the number of
relationships in the graph, relative to many (but not all) uses of the
anchor pattern. I am inclined to think that's not too bad a problem,
however, because containment is only needed in relatively few places, and
because I anticipate that most containing classes don't need to contain
very many other classes. The increase would be less than two relationships
per contained class in the worst case, and zero in the best case.
Any takers?
John
--
You received this message because you are subscribed to the Google Groups
Puppet Users group.
To view this discussion on the web visit
https://groups.google.com/d/msg/puppet-users/-/IYFT8_KB7_UJ.
To post to this group, send email to puppet-users@googlegroups.com.
To unsubscribe from this group, send email to
puppet-users+unsubscr...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/puppet-users?hl=en.
Re: [Puppet Users] yum.puppetlabs.com rsync problem
On Mon, Oct 15, 2012 at 11:45 PM, Andrey Kononov zloyka...@gmail.com wrote: rsync to yum.puppetlabs.com didn`t work now =( 22/tcp open ssh 25/tcp closed smtp 80/tcp open http 443/tcp open https 631/tcp closed ipp /tcp closed unknown 5666/tcp open nrpe It seems to be working for me. Perhaps it was just an intermittent issue? jeff@puppetmaster:~$ *rsync rsync://yum.puppetlabs.com/packages* drwxrwsr-x4096 2012/10/11 20:07:19 . -rw-rw-r--3511 2012/10/11 18:04:44 Rakefile drwxrwsr-x4096 2012/10/10 17:51:41 apt-enterprise drwxrwsr-x4096 2012/10/12 06:44:18 apt drwxrwxr-x 20480 2012/10/12 06:17:48 incoming drwxr-xr-x4096 2012/05/16 21:52:29 tasks drwxrwsr-x4096 2012/05/15 16:33:43 yum-enterprise drwxrwsr-x4096 2012/09/28 18:45:39 yum -Jeff -- You received this message because you are subscribed to the Google Groups Puppet Users group. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
[Puppet Users] How to trace puppet execution to troubleshoot odd problems
What is the best way to get a full trace of what puppet is doing, with variable dumps and state, to troubleshoot problems like where a user's password keeps getting reset every time puppet runs even though the password hashes match? I want to hack out a way to get HPUX working like we have our Linux environments but need to be able to track down right where a problem is coming in so I can adjust / override / submit changes to the code base. Thanks! -- You received this message because you are subscribed to the Google Groups Puppet Users group. To view this discussion on the web visit https://groups.google.com/d/msg/puppet-users/-/yspr-yxcnuEJ. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
[Puppet Users] Tracing puppet processing for HP-UX user passwords
I have finally gotten Ruby 1.8.7 to compile properly with SSL libraries on HPUX so now I am fighting how to get puppet to manage passwords. First I modified the hpux.rb under the providers/user directory to have it use /usr/sam/lbin/useradd.sam to add users and that worked great right off the bat. The problem I am having now is that each time puppet runs it does not seem to be getting the password hash for any user defined on the puppet master box and thus every run is changing the password. Digging into the user.rb under the types directory shows in :password that the value is absent so it goes on to change it. I have run puppet in verbose mode, set RUBYOPT=-d and a wide range of other things to debug the execution but I have no idea what more to do here since I am not a Ruby coder or know the best ways to troubleshoot the flow of the code in Puppet. We can deal with having it doing all the password resets on the users but seems like it will be a lot of changes to password where there really is not an issue. Since I am trying to get this working across the board for HPUX I wanted to try to hack out the fix myself but have hit a wall. Pertinent Details for the host on which I am finding these issues: architecture = ia64 facterversion = 1.6.12 fqdn = cadcpuppet hardwaremodel = ia64 hostname = cadcpuppet id = root interfaces = lan0,lo0 is_virtual = true kernel = HP-UX kernelmajversion = B.11 kernelrelease = B.11.31 kernelversion = B.11.31 operatingsystem = HP-UX operatingsystemrelease = B.11.31 osfamily = HP-UX path = /usr/sbin:/usr/bin:/usr/ccs/bin:/usr/contrib/bin:/usr/contrib/Q4/bin:/opt/perl/bin:/opt/gvsd/bin:/opt/ipf/bin:/opt/nettladm/bin:/opt/fcms/bin:/opt/wbem/bin:/opt/wbem/sbin:/opt/sas/bin:/opt/graphics/common/bin:/opt/atok/bin:/usr/bin/X11:/usr/contrib/bin/X11:/opt/sec_mgmt/bastille/bin:/opt/caliper/bin:/opt/drd/bin:/opt/dsau/bin:/opt/dsau/sbin:/opt/resmon/bin:/opt/firefox:/opt/gnome/bin:/opt/perf/bin:/opt/propplus/bin:/usr/contrib/kwdb/bin:/opt/perl_32/bin:/opt/prm/bin:/opt/sfm/bin:/opt/swm/bin:/opt/sec_mgmt/spc/bin:/opt/ssh/bin:/opt/swa/bin:/opt/hpsmh/bin:/opt/thunderbird:/opt/sentinel/bin:/opt/langtools/bin:/opt/wlm/bin:/opt/gwlm/bin:/sbin:/home/root:/usr/local/bin ps = ps -ef puppetversion = 2.7.19 rubysitedir = /usr/local/lib/hpux32/ruby/site_ruby/1.8 rubyversion = 1.8.7 uptime = 6 days uptime_days = 6 uptime_hours = 164 uptime_seconds = 593014 virtual = hpvm -- You received this message because you are subscribed to the Google Groups Puppet Users group. To view this discussion on the web visit https://groups.google.com/d/msg/puppet-users/-/VUF2tqWzNWoJ. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
Re: [Puppet Users] PHP modules
On 10/16/2012 04:42 PM, Brent Clark wrote: Good day. I would like to ask. Does anyone have an elegant way to manage the PHP modules that are not available via a repository, and that can only be installed via pecl or pear. Ideally, it would actually be better if Debians dh-make-(pear|pecl) was actually used in the manifest. Github is not is not able to help. Build your own packages and distribute them via internal apt repo. It's not that hard... -- You received this message because you are subscribed to the Google Groups Puppet Users group. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
Re: [Puppet Users] Setting up Java on Windows
On Tue, Oct 16, 2012 at 6:14 AM, Matthew Burgess
matthew.2.burg...@googlemail.com wrote:
On Tue, Oct 16, 2012 at 5:45 AM, wukong wicked.wuk...@gmail.com wrote:
Also is there a way to do this as silent install?
Here is my manifest.
package {java :
ensure = installed,
provider = windows,
source = C:/Manifests/installables/jdk-6u35-windows-i586.exe,
}
Launching that .exe will always give you a GUI, unless you provide
some additional arguments. To install the JRE silently, see the
following:
http://www.oracle.com/technetwork/java/javase/silent-136552.html#running.
Regards,
Matt.
--
You received this message because you are subscribed to the Google Groups
Puppet Users group.
To post to this group, send email to puppet-users@googlegroups.com.
To unsubscribe from this group, send email to
puppet-users+unsubscr...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/puppet-users?hl=en.
Having done this recently, I'd recommend something like the following:
package { 'Java(TM) 7 Update 5':
ensure = installed,
source = 'path/jre-7u5-windows-i586.exe',
install_options = [
'/s', '/v/qn ADDLOCAL=jrecore REBOOT=Suppress JAVAUPDATE=0'
]
}
Note the name of the package needs to match 'DisplayName' as specified
in the registry and displayed in Add/Remove programs.
As for customizing the PATH, this is filed as
https://projects.puppetlabs.com/issues/16048 and there's some
background information here
https://groups.google.com/forum/?fromgroups=#!topic/puppet-users/68ebrAEUzgQ
Josh
--
Josh Cooper
Developer, Puppet Labs
--
You received this message because you are subscribed to the Google Groups
Puppet Users group.
To post to this group, send email to puppet-users@googlegroups.com.
To unsubscribe from this group, send email to
puppet-users+unsubscr...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/puppet-users?hl=en.
Re: [Puppet Users] How to trace puppet execution to troubleshoot odd problems
On Tue, Oct 16, 2012 at 10:48 AM, HPUX_PUPPET dride...@gmail.com wrote: What is the best way to get a full trace of what puppet is doing, with variable dumps and state, to troubleshoot problems like where a user's password keeps getting reset every time puppet runs even though the password hashes match? In these situations I try and run both the master and the agent with the following options turned on: --verbose --debug --trace --evaltrace I want to hack out a way to get HPUX working like we have our Linux environments but need to be able to track down right where a problem is coming in so I can adjust / override / submit changes to the code base. With the --debug flag, providers will print out the exact command they're executing to modify the system. --noop is also useful to see what commands Puppet _would_ execute without actually executing them. -Jeff -- You received this message because you are subscribed to the Google Groups Puppet Users group. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
[Puppet Users] Custom Ruby
Is it possible to tell Puppet (and Facter for that matter) to use a custom Ruby that exists alongside the Ruby 1.8.7 that comes with CentOS 5 without converting the entire host to only use the custom version? I'm in an environment where we're stuck on CentOS 5 for some time and it's ancient Ruby. We do though have a custom packaging of Ruby 1.9 installed on all machines with separate binaries, i.e. /usr/bin/my_ruby, /usr/bin/my_irb that we use when we need Ruby. I'm looking for some config or startup option that says where to find Ruby. So far I don't see anything in the init scripts or config options. Solutions such as linking /usr/bin/ruby to /usr/bin/my_ruby are not an option but rebuilding an RPM is. Thanks! -- You received this message because you are subscribed to the Google Groups Puppet Users group. To view this discussion on the web visit https://groups.google.com/d/msg/puppet-users/-/MdQ1apUpOxQJ. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
[Puppet Users] Re: Custom Ruby
BTW, I want to do this so that I can upgrade to Puppet 3 On Tuesday, October 16, 2012 12:15:30 PM UTC-7, Jamie wrote: Is it possible to tell Puppet (and Facter for that matter) to use a custom Ruby that exists alongside the Ruby 1.8.7 that comes with CentOS 5 without converting the entire host to only use the custom version? I'm in an environment where we're stuck on CentOS 5 for some time and it's ancient Ruby. We do though have a custom packaging of Ruby 1.9 installed on all machines with separate binaries, i.e. /usr/bin/my_ruby, /usr/bin/my_irb that we use when we need Ruby. I'm looking for some config or startup option that says where to find Ruby. So far I don't see anything in the init scripts or config options. Solutions such as linking /usr/bin/ruby to /usr/bin/my_ruby are not an option but rebuilding an RPM is. Thanks! -- You received this message because you are subscribed to the Google Groups Puppet Users group. To view this discussion on the web visit https://groups.google.com/d/msg/puppet-users/-/SeIFJGKJilsJ. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
[Puppet Users] puppet module and no internet connection
Hi, I managed to get a puppet module on a server and tried to install it but I got the following and am now lost on how to get round this: puppet module install puppetlabs-xinetd-1.1.0 Preparing to install into /etc/puppet/modules ... Downloading from http://forge.puppetlabs.com ... Error: Error: Could not connect to http://forge.puppetlabs.com There was a network communications problem Check your network connection and try again The server doesn't have an internet connection so how do I install this? Thanks in advance for any help. Dan -- You received this message because you are subscribed to the Google Groups Puppet Users group. To view this discussion on the web visit https://groups.google.com/d/msg/puppet-users/-/NyDv8k7gDQIJ. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
[Puppet Users] Puppet dashboard over passenger: The environment must be purely alphanumeric, not ''
Hello, I installed puppet-server, passenger and dashboard. I have tested every step in between to make sure all is running well and not errors are carrying forward. Dashboard worked well with my test in WEBrick. After configure the dashboard with passenger I got the message” The environment must be purely alphanumeric, not ''” instead of the dashboard. I verified the configuration files and all look to be in place, but I should be missing something. This is my current environment: CentOS release 6.3 x86_64 puppet-2.7.19-1.el6.noarch puppet-server-2.7.19-1.el6.noarch puppet-dashboard-1.2.12-1.el6.noarch *** LOCAL GEMS *** activemodel (3.2.8) activerecord (3.2.8) activesupport (3.2.8) arel (3.0.2) builder (3.0.3) daemon_controller (1.0.0) fastthread (1.0.7) i18n (0.6.1) json (1.4.6) multi_json (1.3.6) passenger (3.0.17) rack (1.4.1) rake (0.9.2.2, 0.8.7) tzinfo (0.3.33) # rpm -qa | grep ruby ruby-devel-1.8.7.352-7.el6_2.x86_64 ruby-mysql-2.8.2-1.el6.x86_64 ruby-libs-1.8.7.352-7.el6_2.x86_64 ruby-irb-1.8.7.352-7.el6_2.x86_64 rubygems-1.3.7-1.el6.noarch ruby-shadow-1.4.1-13.el6.x86_64 libselinux-ruby-2.0.94-5.3.el6.x86_64 ruby-augeas-0.4.1-1.el6.x86_64 rubygem-json-1.4.6-1.el6.x86_64 rubygem-rake-0.8.7-2.1.el6.noarch ruby-ri-1.8.7.352-7.el6_2.x86_64 ruby-1.8.7.352-7.el6_2.x86_64 ruby-rdoc-1.8.7.352-7.el6_2.x86_64 rpm -qa | grep mysql ruby-mysql-2.8.2-1.el6.x86_64 mysql-5.1.61-4.el6.x86_64 mysql-devel-5.1.61-4.el6.x86_64 mysql-libs-5.1.61-4.el6.x86_64 mysql-server-5.1.61-4.el6.x86_64 Any help will be welcome. Thanks, Andrew -- You received this message because you are subscribed to the Google Groups Puppet Users group. To view this discussion on the web visit https://groups.google.com/d/msg/puppet-users/-/EuSF65nrsroJ. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
Re: [Puppet Users] Custom Ruby
You can rebuild puppet and/or facter to use the ruby you have available. It should only require modifying a couple of lines in the spec file. You'd need to modify the ruby dependencies and the rubylib. On Tue, Oct 16, 2012 at 12:15 PM, Jamie gojamieg...@gmail.com wrote: Is it possible to tell Puppet (and Facter for that matter) to use a custom Ruby that exists alongside the Ruby 1.8.7 that comes with CentOS 5 without converting the entire host to only use the custom version? I'm in an environment where we're stuck on CentOS 5 for some time and it's ancient Ruby. We do though have a custom packaging of Ruby 1.9 installed on all machines with separate binaries, i.e. /usr/bin/my_ruby, /usr/bin/my_irb that we use when we need Ruby. I'm looking for some config or startup option that says where to find Ruby. So far I don't see anything in the init scripts or config options. Solutions such as linking /usr/bin/ruby to /usr/bin/my_ruby are not an option but rebuilding an RPM is. Thanks! -- You received this message because you are subscribed to the Google Groups Puppet Users group. To view this discussion on the web visit https://groups.google.com/d/msg/puppet-users/-/MdQ1apUpOxQJ. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en. -- Matthaus Owens Release Manager, Puppet Labs -- You received this message because you are subscribed to the Google Groups Puppet Users group. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
[Puppet Users] puppet windows importing manifests twice when using puppet apply
I am trying to create a masterless local windows installer and I have run
into an issue where puppet is importing my manfiests multiple times
creating multiple declaration errors.
You may notice the importing occurs twice but with different directory
layouts. One import is via puppet while the other has something to do with
the modulepath I declared.
Any ideas?
Corey
Here is what I am doing (somewhat abbreviated):
ruby -S -- puppet apply -v --debug --noop --modulepath=dir/puppetmodules
-e include myclass
debug: importing 'C:\devops\trunk/puppetmodules/myclass/manifests/init.pp'
in environment production
debug: importing 'C:/devops/trunk/puppetmodules/myclass/manifests/init.pp'
in environment production
debug: Automatically imported myclass from myclass into production
debug: importing 'C:\devops\trunk/puppetmodules/jdk/manifests/init.pp' in
environment production
debug: importing 'C:/devops/trunk/puppetmodules/jdk/manifests/init.pp' in
environment production
debug: Automatically imported jdk from jdk into production
Duplicate declaration: Exec[create_java_home] is already declared in file
C:\devops\trunk/puppetmodules/jdk/manifests/init.pp at line 18; cannot
redeclare at C:/devops/trunk/puppetmodules/jdk/manifests/init.pp:18 on
node hostname_hidden
Here is the myclass (changed name for this email)
class myclass {
class {myclass::install: }
# Now try and use maven to run the build
class {maven::command:
build_dir = c:/${mvn_repo_name},
command = verify -Ptests,
require = [Exec[checksvnrepo],Class[myclass::install]],
}
}
class myclass::install{
if $::operatingsystem == windows {
class {jdk: }
class {svn: }
# since springsource comes with maven lets use the builtin maven package
if $::install_sts == true {
class {sts:
require = Class[jdk]
}
$mvn_dir='C:\Program Files\springsource\apache-maven-3.0.3'
class {maven::setup:
mvn_home = $mvn_dir
}
}
else{
# Install and Setup Maven
class {maven: }
}
}
}
Here is the jdk class being called since its the first class
class jdk{
class {jdk::setup:
require = Class[jdk::install],
java_home = 'C:/Program Files/Java/jdk1.6.0_30',
java_bin = '%JAVA_HOME%/bin',
}
}
class jdk::setup($java_home, $java_bin){
# set the user environment variable for java_home
exec{create_java_home:
command = setx JAVA_HOME \${java_home}\ /M ,
path = 'C:\Windows\System32',
# The unless should really just check if the variable is set for the
specific path
unless = where javac,
}
# set the user environment variable for java_bin
exec{create_java_bin:
command = setx JAVA_BIN \${java_bin}\ /M ,
path = 'C:\Windows\System32',
unless = where javac,
}
}
class jdk::install{
exec{jdkinstaller:
command = ${::installer_dir}/jdk-6u30-windows-i586.exe /s
ADDLOCAL=\ToolsFeature,DemosFeature,SourceFeature\ ,
creates = 'C:\Program Files\Java\jdk1.6.0_30\bin\javac.exe',
logoutput = 'true',
}
}
--
You received this message because you are subscribed to the Google Groups
Puppet Users group.
To view this discussion on the web visit
https://groups.google.com/d/msg/puppet-users/-/CuB0G1TQ9SkJ.
To post to this group, send email to puppet-users@googlegroups.com.
To unsubscribe from this group, send email to
puppet-users+unsubscr...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/puppet-users?hl=en.
Re: [Puppet Users] Custom Ruby
Awesome! I'll try that, thank you. On Tuesday, October 16, 2012 1:27:21 PM UTC-7, Matthaus Litteken wrote: You can rebuild puppet and/or facter to use the ruby you have available. It should only require modifying a couple of lines in the spec file. You'd need to modify the ruby dependencies and the rubylib. On Tue, Oct 16, 2012 at 12:15 PM, Jamie gojam...@gmail.com javascript: wrote: Is it possible to tell Puppet (and Facter for that matter) to use a custom Ruby that exists alongside the Ruby 1.8.7 that comes with CentOS 5 without converting the entire host to only use the custom version? I'm in an environment where we're stuck on CentOS 5 for some time and it's ancient Ruby. We do though have a custom packaging of Ruby 1.9 installed on all machines with separate binaries, i.e. /usr/bin/my_ruby, /usr/bin/my_irb that we use when we need Ruby. I'm looking for some config or startup option that says where to find Ruby. So far I don't see anything in the init scripts or config options. Solutions such as linking /usr/bin/ruby to /usr/bin/my_ruby are not an option but rebuilding an RPM is. Thanks! -- You received this message because you are subscribed to the Google Groups Puppet Users group. To view this discussion on the web visit https://groups.google.com/d/msg/puppet-users/-/MdQ1apUpOxQJ. To post to this group, send email to puppet...@googlegroups.comjavascript:. To unsubscribe from this group, send email to puppet-users...@googlegroups.com javascript:. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en. -- Matthaus Owens Release Manager, Puppet Labs -- You received this message because you are subscribed to the Google Groups Puppet Users group. To view this discussion on the web visit https://groups.google.com/d/msg/puppet-users/-/E1SHrHqtNCgJ. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
[Puppet Users] Re: Setting up Java on Windows
I wasn't aware that you could use the package resource for exe files since
it looked like it was limited to msi files.
I ended up crafting up this solution using exec resource
exec{jdkinstaller:
command = ${::installer_dir}/jdk-6u30-windows-i586.exe /s
ADDLOCAL=\ToolsFeature,DemosFeature,SourceFeature\ ,
creates = 'C:\Program Files\Java\jdk1.6.0_30\bin\javac.exe',
logoutput = 'true',
}
Still struggling to set the Global PATH variable in a nice way.
Corey
On Monday, October 15, 2012 11:45:25 PM UTC-5, wukong wrote:
Hi guys,
I have been trying to bootstrap a windows machine with puppet. I need to
install java and setup the JAVA_HOME and add bin to the path variable of
windows.
I have tried using the package resource with the jdk setup but everytime I
run the manifest puppet opens the setup wizard and waits for me to do
something.
Am I doing something wrong or is this the right behavior? Also is there a
way to do this as silent install?
Here is my manifest.
package {java :
ensure = installed,
provider = windows,
source = C:/Manifests/installables/jdk-6u35-windows-i586.exe,
}
Thank you.
--
You received this message because you are subscribed to the Google Groups
Puppet Users group.
To view this discussion on the web visit
https://groups.google.com/d/msg/puppet-users/-/IJlOOM-dRZUJ.
To post to this group, send email to puppet-users@googlegroups.com.
To unsubscribe from this group, send email to
puppet-users+unsubscr...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/puppet-users?hl=en.
[Puppet Users] Re: Unable to create the rrd graph
Can anybody reply on how to resolve this issue On Tuesday, October 16, 2012 9:45:49 AM UTC-4, Nishant Jain wrote: Hello Everybody, I have been trying to use the rrdgraph reporting feature available in puppet. After performing all the instruction maintained on the site http://docs.puppetlabs.com/references/latest/report.html , the rrd directory for the agents get created but they are empty. When i check the log files in /var/log/messages, I get to see the following error Failed to set owner to '0': Operation not permitted - /var/lib/puppet/rrd/ftldwshost174.wsdev.citrix.com RRD library is missing; cannot store metrics I have installed the RRD lib and tools using the following commands :- yum install rrdtool-ruby I tried installing from the source from the following url:http://rubyforge.org/projects/rubyrrdtool/ But it tends to give an error in make Can anybody tell me how to resolve this issue. Thanks, Nishant -- You received this message because you are subscribed to the Google Groups Puppet Users group. To view this discussion on the web visit https://groups.google.com/d/msg/puppet-users/-/71ImPRMSM_UJ. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
Re: [Puppet Users] puppet windows importing manifests twice when using puppet apply
Hi Corey, On Tue, Oct 16, 2012 at 1:34 PM, Corey Osman co...@logicminds.biz wrote: I am trying to create a masterless local windows installer and I have run into an issue where puppet is importing my manfiests multiple times creating multiple declaration errors. You may notice the importing occurs twice but with different directory layouts. One import is via puppet while the other has something to do with the modulepath I declared. Any ideas? Corey Here is what I am doing (somewhat abbreviated): ruby -S -- puppet apply -v --debug --noop --modulepath=dir/puppetmodules -e include myclass debug: importing 'C:\devops\trunk/puppetmodules/myclass/manifests/init.pp' in environment production debug: importing 'C:/devops/trunk/puppetmodules/myclass/manifests/init.pp' in environment production I'm pretty sure this is https://projects.puppetlabs.com/issues/13055 Try specifying the modulepath as `--modulepath=C:/devops/trunk/puppetmodules` The issue here is that the win32-dir gem patches Dir.pwd, causing it to return a path with backslashes. This confuses the autoloader in 2.7.x. However, this should not be an issue in 3.0. If so, please let me know. Josh -- Josh Cooper Developer, Puppet Labs -- You received this message because you are subscribed to the Google Groups Puppet Users group. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
Re: [Puppet Users] puppet module and no internet connection
Hi Dan, If you can't connect to Forge from your node and using the http_proxy configuration option isn't viable, you can download the tarball directly from the Forge website and install it with the puppet module tool. First, download the tarball from the website: http://forge.puppetlabs.com/puppetlabs/xinetd Get that tarball onto your node somehow and then run the `puppet module install` command against it. The --ignore-dependencies flag instructs the tool to ignore the Forge (which is normally consulted for dependency lookup) and just install the module. If the module does have dependencies (listed on the web page), you'll need to install those as well. Example: puppet module install puppetlabs-xinetd-1.1.0.tar.gz --ignore-dependencies Hope that helps! --Ryan On Tue, Oct 16, 2012 at 12:32 PM, Dan dan.tr...@gmail.com wrote: Hi, I managed to get a puppet module on a server and tried to install it but I got the following and am now lost on how to get round this: puppet module install puppetlabs-xinetd-1.1.0 Preparing to install into /etc/puppet/modules ... Downloading from http://forge.puppetlabs.com ... Error: Error: Could not connect to http://forge.puppetlabs.com There was a network communications problem Check your network connection and try again The server doesn't have an internet connection so how do I install this? Thanks in advance for any help. Dan -- You received this message because you are subscribed to the Google Groups Puppet Users group. To view this discussion on the web visit https://groups.google.com/d/msg/puppet-users/-/NyDv8k7gDQIJ. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en. -- You received this message because you are subscribed to the Google Groups Puppet Users group. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
Re: [Puppet Users] Custom Ruby
On 10/17/2012 01:02 AM, Craig White wrote: The puppet server should probably be running on something newer than CentOS 5.x (a vm perhaps?) Can you run 3.0 server with 2.7 clients? -- Jakov Sosic www.srce.unizg.hr -- You received this message because you are subscribed to the Google Groups Puppet Users group. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
[Puppet Users] Access to the Puppet open source tarballs
Hi all There has been some discussion lately on the list about a change we made to the website. The change removed the link that allows you to bypass providing your email when downloading tarballs on this page: http://info.puppetlabs.com/download-puppet-open-source In light of that discussion we've decided to restore that link. We’re sorry that the change was made without adequately consulting the community. We’re committed to being as transparent and open as possible about our business and how it interacts with the open source community. In this case we screwed up and didn’t do a good job of that. We plan to learn from that mistake and ensure it doesn’t happen again. We also want to make clear that this page is just one of the many ways you can get access to Puppet. The others include: 1. Github (for example https://github.com/puppetlabs/puppet/zipball/3.0.0) 2. Packages (http://docs.puppetlabs.com/guides/puppetlabs_package_repositories.html) 3. RubyGems (gem install puppet) 4. From your favorite distribution or repository service such as EPEL. You can see other methods here: http://puppetlabs.com/puppet/puppet-open-source/. A number of other comments were made about Puppet Labs commitment to open source that we also want to address. Puppet Labs remains committed to open source. We could not have gotten where we are today without the support, enthusiasm and passion of the open source community. Many of us at Puppet Labs have a long history of involvement in open source and a strong commitment to the success of the open source movement. Puppet continues to be released under the Apache 2.0 license. Apache 2.0 is a free and open source license shared by a huge number of projects including everything released by the Apache Foundation, for example the Apache web server itself. The license provides you with free and open source software as well as allows us to ensure Puppet becomes more ubiquitous and widespread across the industry. Puppet Labs has also continued to release and maintain Puppet, Facter, mcollective, Hiera, and a number of other tools as open source tools. We also plan to release new open source tools (as we did twice this year with PuppetDB and Razor). We have no plans to stop doing this. Indeed we've committed significant engineering resources to all these projects (2447 commits to Puppet alone just this year!) and will continue to build on this. Puppet Labs is a business though. The reason we've been able to commit so much to developing Puppet and its ecosystem is because we've sold Puppet Enterprise to customers. Many in our open source community may never buy Puppet Enterprise: it may not suit your price point, your business or you may not have need of the features it offers. But every license of Puppet Enterprise we sell pays engineers and allows us to grow and develop the Puppet platform. As a result you’ll see a strong focus on Puppet Enterprise on our website and in our marketing. As we mentioned above we’re committed to being as transparent and open as possible about our business and how it interacts with the open source community. Please feel free to hold us to that commitment. We can’t always guarantee we’ll be able to make everyone in the community happy but we will be as clear and open as we can about the changes and decisions we make. Please feel free to reach out to me, Dawn Foster or anyone else at Puppet Labs if you have any questions or concerns. Kind Regards James Turnbull -- James Turnbull Puppet Labs 1-503-734-8571 To schedule a meeting with me: http://doodle.com/jamtur01 -- You received this message because you are subscribed to the Google Groups Puppet Users group. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
Re: [Puppet Users] Custom Ruby
On Tue, Oct 16, 2012 at 4:37 PM, Jakov Sosic jso...@srce.hr wrote: On 10/17/2012 01:02 AM, Craig White wrote: The puppet server should probably be running on something newer than CentOS 5.x (a vm perhaps?) Can you run 3.0 server with 2.7 clients? Yes. -- Jakov Sosic www.srce.unizg.hr -- You received this message because you are subscribed to the Google Groups Puppet Users group. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en. -- You received this message because you are subscribed to the Google Groups Puppet Users group. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
[Puppet Users] Re: Access to the Puppet open source tarballs
James, Thanks for the discussions and speedy resolution. Much appreciated. All the best, Adam Kosmin -- You received this message because you are subscribed to the Google Groups Puppet Users group. To view this discussion on the web visit https://groups.google.com/d/msg/puppet-users/-/8Fc-KWjO5vEJ. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
[Puppet Users] Custom Provider not being loaded
I am attempting to use Chocolatey ( http://chocolatey.org ) as a package
provider for Puppet, but I continually get this error when running the
Puppet agent on my Win 7 box.
info: Caching catalog for ldap-windows-vm.localdomain
err: Failed to apply catalog: Parameter provider failed: Invalid package provide
r 'chocolatey' at /etc/puppet/environments/beta/modules/stormaas/manifests/init.
pp:9
I've placed this module, https://github.com/rismoney/puppet-chocolatey, in
my modules/ directory on my Puppet master, I have pluginsync set to true on
both the master and slave - and am already using a few imported types that
are working with the same setup.
Calling the package with:
package { 'stormaas':
ensure = installed,
provider = 'chocolatey',
}
I've tried with both the provider set to 'chocolatey' and chocolatey in
case it was a quotes issue, and restarted both the master and slave on
every change and ran the agent multiple times to ensure it was refreshed.
All other aspects of the module that is being applied to the machine works
- just not this.
It seems there *are* modules being loaded
into C:\ProgramData\PuppetLabs\puppet\var\lib\puppet\provider , including
this guy which I installed yesterday:
http://forge.puppetlabs.com/puppetlabs/dism , but this is not one of them.
Any thoughts?
--
You received this message because you are subscribed to the Google Groups
Puppet Users group.
To view this discussion on the web visit
https://groups.google.com/d/msg/puppet-users/-/2p1wylYT1pwJ.
To post to this group, send email to puppet-users@googlegroups.com.
To unsubscribe from this group, send email to
puppet-users+unsubscr...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/puppet-users?hl=en.
Re: [Puppet Users] Custom Ruby
On Tue, Oct 16, 2012 at 4:50 PM, Michael Stahnke stah...@puppetlabs.com wrote: On Tue, Oct 16, 2012 at 4:37 PM, Jakov Sosic jso...@srce.hr wrote: On 10/17/2012 01:02 AM, Craig White wrote: The puppet server should probably be running on something newer than CentOS 5.x (a vm perhaps?) Can you run 3.0 server with 2.7 clients? Yes. Also http://docs.puppetlabs.com/guides/upgrading.html#always-upgrade-the-puppet-master-first Josh -- Josh Cooper Developer, Puppet Labs -- You received this message because you are subscribed to the Google Groups Puppet Users group. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
[Puppet Users] Re: Custom Provider not being loaded
Got it.
Not sure if this is a bug, if I was placing it in the wrong scope and
that's just bad practice - but:
The provider needed to be in ` /etc/puppet/modules ` while I had it in `
/etc/puppet/environments/beta/modules ` (With the rest of the items in my
beta environment).
Would still love to hear input on this though :-)
TJ
On Tuesday, October 16, 2012 4:32:55 PM UTC-7, Thomas Biddle wrote:
I am attempting to use Chocolatey ( http://chocolatey.org ) as a package
provider for Puppet, but I continually get this error when running the
Puppet agent on my Win 7 box.
info: Caching catalog for ldap-windows-vm.localdomain
err: Failed to apply catalog: Parameter provider failed: Invalid package
provide
r 'chocolatey' at
/etc/puppet/environments/beta/modules/stormaas/manifests/init.
pp:9
I've placed this module, https://github.com/rismoney/puppet-chocolatey,
in my modules/ directory on my Puppet master, I have pluginsync set to true
on both the master and slave - and am already using a few imported types
that are working with the same setup.
Calling the package with:
package { 'stormaas':
ensure = installed,
provider = 'chocolatey',
}
I've tried with both the provider set to 'chocolatey' and chocolatey in
case it was a quotes issue, and restarted both the master and slave on
every change and ran the agent multiple times to ensure it was refreshed.
All other aspects of the module that is being applied to the machine works
- just not this.
It seems there *are* modules being loaded
into C:\ProgramData\PuppetLabs\puppet\var\lib\puppet\provider , including
this guy which I installed yesterday:
http://forge.puppetlabs.com/puppetlabs/dism , but this is not one of
them.
Any thoughts?
--
You received this message because you are subscribed to the Google Groups
Puppet Users group.
To view this discussion on the web visit
https://groups.google.com/d/msg/puppet-users/-/_9eTNb_1oDgJ.
To post to this group, send email to puppet-users@googlegroups.com.
To unsubscribe from this group, send email to
puppet-users+unsubscr...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/puppet-users?hl=en.
Re: [Puppet Users] Re: Custom Provider not being loaded
Hi TJ, On Tue, Oct 16, 2012 at 7:00 PM, Thomas Biddle urins...@gmail.com wrote: Got it. Not sure if this is a bug, if I was placing it in the wrong scope and that's just bad practice - but: The provider needed to be in ` /etc/puppet/modules ` while I had it in ` /etc/puppet/environments/beta/modules ` (With the rest of the items in my beta environment). Aw, crap. I think we might have an open bug about this somewhere. Had you run a puppet agent with pluginsync on the master? I think there's something about having the master syncing plugins to its own libdir from an environment before it'll serve that to agents. I'm hand-waving most of this because I can't find the ticket I'm thinking of. Anyhow, if you try that.. can you sync the provider to a windows agent from your environment? Would still love to hear input on this though :-) TJ -- You received this message because you are subscribed to the Google Groups Puppet Users group. To post to this group, send email to puppet-users@googlegroups.com. To unsubscribe from this group, send email to puppet-users+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en.
