On 01/09/2011 09:11 PM, Stefan Schulte wrote:
> On Fri, Jan 07, 2011 at 09:41:39PM +, R.I.Pienaar wrote:
>> (forced to top post due to your html email)
>>
>> You cant avoid the scoping issues cos its broken. You can work around them
>> though.
>>
>
> To be honest I always found the current
On 01/11/2011 07:59 PM, Robin Lee Powell wrote:
> On Tue, Jan 11, 2011 at 10:41:13AM -0800, Daniel Pittman wrote:
>> The other thing I keep in mind is that using a 'define' that wraps
>> some file and exec operations *is* keeping with the spirit of that
>> statement, if not the letter: sure, it def
Hi, I have a simple use case, but can't get the dependency resolution
to work on first run. This is not critical, but I want to understand
what's going on. Here's the szenario:
* puppet --version: 2.6.4
* on the bare system, I have vim 6.4 installed (no vim-base)
* I first want to upgrade to the l
I sometimes get the same error on puppet 2.6.4 and don't know the
cause. Restarting the agent helps, but I would welcome a solution to
this.
On Jan 12, 1:25 am, Mohamed Lrhazi wrote:
> What does this error mean or what does it indicate?
>
> puppet-agent: Could not retrieve catalog from remote s
On 01/12/2011 11:30 AM, Sven Sporer wrote:
> Hi, I have a simple use case, but can't get the dependency resolution
> to work on first run. This is not critical, but I want to understand
> what's going on. Here's the szenario:
>
> * puppet --version: 2.6.4
> * on the bare system, I have vim 6.4 ins
On Jan 12, 2011, at 2:30 AM, Sven Sporer wrote:
> Hi, I have a simple use case, but can't get the dependency resolution
> to work on first run. This is not critical, but I want to understand
> what's going on. Here's the szenario:
>
> * puppet --version: 2.6.4
> * on the bare system, I have vim
On Jan 12, 12:27 pm, Patrick wrote:
> On Jan 12, 2011, at 2:30 AM, Sven Sporer wrote:
>
>
>
>
>
>
>
>
>
> > Hi, I have a simple use case, but can't get the dependency resolution
> > to work on first run. This is not critical, but I want to understand
> > what's going on. Here's the szenario:
>
> >
On Wed, Jan 12, 2011 at 03:43:42AM -0800, Sven Sporer wrote:
> > This makes no sense to me that the last doesn't always work.
> > Someone might come up with an answer, but logs with "--verbose
> > --debug" would probably help. Even if you sanitize it with "grep
> > vim" it would probably help a l
Stig Sandbeck Mathisen writes:
> ,
> | node *.desktops.example.org {
> | case $server_role {
> | /:desktop:/ { include my::desktop::class }
> | /:mission_critical:/ { include corp::absolutely::vital:class }
> | /:whohooo:/ { include rubber::ducky }
> | }
> | }
> > What do you mean by "the last"? Everything is installed as intended
> > after the third run. Sometimes, even after the second run.
>
> That's what he means. Why this inconsistency?
>
> > So I'm not
> > really sure what causes this inconsistency, but it has something to do
> > with yum dependen
On Jan 12, 7:09 am, Sven Sporer wrote:
[...]
> The initial problem still stands: why do I need a second run? What
> changes in between for the client? Or should I blame it on this
> version of yum? ;)
>
> Here's the log output:http://pastie.org/1452205
The log shows Puppet attempting to install
Hello Patrick,
I have the puppet-dashboard up and running and get the gui. What I am
seeing there is that none
of my clients(nodes) are reporting. When I click on nodes the dashboard
lists all of my nodes but
also states each has not reported. What I am trying to get working is
that the puppet cl
On Tue, Jan 11, 2011 at 12:07:30PM -0800, Robin Lee Powell wrote:
> On Tue, Jan 11, 2011 at 05:20:38AM -0800, Al @ Lab42 wrote:
> >
> > You can build a file based on different "fragments" at least in
> > 2 ways:
> >
> > - When you specify an array of templates , when using the
> > content => argume
these are manual step to be taken so one can install puppet from the
backport repo
so not sure what you ref. too ? the order of step 1, 2 and 3 does not
matter but need to be done before 4 (apt-get update)
On Jan 11, 2:25 pm, Patrick wrote:
> On Jan 11, 2011, at 8:52 AM, Luc Suryo wrote:
>
> > 1
I get this very often... I only have a 4 or 5 hosts setup right now...
with version 2.6.3 everywhere.
Sometimes I run "puppetd --verbose --onetime --no-daemonize", get the
error, then run again and not get it.
Sometimes I run it dozens of times in a row and get it all the
time it's driving me
Hey.
You need to do two things, both of which should be in the install
instructions - so please file a bug if they are not, or are unclear.
The first is to turn on the report option on your clients so they send data
to the master. (Kiterally, report=true in the agent config)
The second is to se
Are you using Passenger? If you don't know, the answer is probably no. If
not, do this:
Shutdown your puppet server. This might work:
service puppetmaster stop
Start it in debug:
puppetmasterd --verbose --debug --no-daemonize --trace
Then watch the output on the console for hints while runni
Except that, knowing this list, someone will automate it with puppet.
On Jan 12, 2011, at 8:14 AM, Luc Suryo wrote:
> these are manual step to be taken so one can install puppet from the
> backport repo
> so not sure what you ref. too ? the order of step 1, 2 and 3 does not
> matter but need to
Mohamed Lrhazi wrote:
Thanks a lot guys... but how do I do this, both silencing reporting
and overriding noop mode, for a resource other than "exec"... mine is:
file {
"/etc/puppet/facts.yaml":
ensure => file,
content => inline_template("<%= scope.to_hash.reject { |k,v| !(
k.is_
Hi Daniel,
I have already done the setup which you mention below. I followed the
puppet-dashboard configuration guide to the tee,
which is why I'm surprised that this feature is not working. I have
verified that the puppet.conf entries on the client and
master are correct. I'm wondering if
Thanks you so much Bellman, the loglevel seems to be missing corner stone :)
Mohamed.
On Wed, Jan 12, 2011 at 2:16 PM, Thomas Bellman wrote:
> Mohamed Lrhazi wrote:
>
>> Thanks a lot guys... but how do I do this, both silencing reporting
>> and overriding noop mode, for a resource other than "ex
Felix Frank wrote:
That's why I originally asked for pastes of configs, manifests *and*
filesystem listings (not to the list preferably, use pastebin services).
I strongly disagree! Having things on a website instead of in
the actual mail I'm reading and possibly responding to sucks.
It's one
# cat iddmi.pp
class iddmi {
# define pkg variable
$pkg = $hardwareisa ? {
"sparc" => "Iddmi_SunOS_sparc.pkg",
"i386" => "Iddmi_SunOS_i386.pkg",
default => undef,
}
$ftpsystems = "ftp.example.net/download"
# define path deafult fo
Thanks Patrick. I see no clues at all:
$sudo /opt/ruby-enterprise/bin/ruby /opt/ruby-enterprise/bin/puppet
master --no-daemonize --debug --verbose --trace
...
notice: Starting Puppet master version 2.6.3
...
info: Expiring the node cache of akubra.example.com
info: Not using expired node for akub
hey,
- Original Message -
> Thanks you so much Bellman, the loglevel seems to be missing corner
> stone :)
This is great, I never know the loglevel meta param did this.
Updated the mcollective wiki to reflect this :)
>
> Mohamed.
>
> On Wed, Jan 12, 2011 at 2:16 PM, Thomas Bellman
>
Hello everyone,
I'm new to the Puppet world and I was looking for a little beginning
help...
I'd like to create a banner for anyone logging into a list of servers
to see (i.e. Unauthorized access prohibited, you're being watched,
etc...).
I'd like to base it off the /etc/issue and /etc/issue.net
I get it too. It's not related to the compilation of the modules but I figure
it's possibly related to the webrick server. I don't see it on my
passenger/apache set up(which I run on another master).
For what it is worth this is good reading for checking the validity of of your
modules before y
On Jan 12, 2011, at 11:37 AM, wrote:
> Hi Daniel,
>
> I have already done the setup which you mention below. I followed the
> puppet-dashboard configuration guide to the tee,
> which is why I’m surprised that this feature is not working. I have verified
> that the puppet.conf entries on the
I have the following use case, which I think is a very valid use case,
but puppet is throwing an error. Puppet's implementation of how
parameterized classes work with include/require breaks the
*declarative* nature of Puppet. I believe this is the same bug as:
http://projects.puppetlabs.com/issu
I'm trying to setup my puppetmaster to use apache/mongrel. I've
followed the guide at
http://projects.puppetlabs.com/projects/1/wiki/Using_Mongrel
for doing this.
My puppetmaster starts fine. When I have a client hit the
puppetmaster I get the following.
>From master:
err: uninitialized const
On 12 January 2011 19:08, Zman112233 wrote:
> I'd like to create a banner for anyone logging into a list of servers
> to see (i.e. Unauthorized access prohibited, you're being watched,
> etc...).
I have a "prettymotd" module. The init.pp reads:
class prettymotd {
file { "/etc/motd":
I forgot to mention, the other annoying issue is that the error
message points to the file and line number of the parameterized class
declaration, NOT the require statement. Enabling verbose/debug output
doesn't help in finding the offending require statement. I basically
had to figure out for my
that would be a catch 22 :)
the instruction was an answer to how to install puppet on a Lucid
server, there is no puppet installed yet!
So how will one automate 'with' puppet while puppet is not installed?
On Jan 12, 10:49 am, Patrick wrote:
> Except that, knowing this list, someone will auto
Hello,
I'm running 0.25.4 on Ubuntu 10.04 LTS. Recently when I start
puppetmaster the 8140 port opens but if any client machine connects to
it (ie telnet localhost 8140), the connection immediately dies.
Network wise I see a syn, syn-ack, ack then a fin-ack in both
directions and then a final ack
Well, your instructions also can work for upgrading puppet with a few small
changes. Might even work without changes.
On Jan 12, 2011, at 2:54 PM, Luc Suryo wrote:
> that would be a catch 22 :)
>
> the instruction was an answer to how to install puppet on a Lucid
> server, there is no puppet i
On Wed, Jan 12, 2011 at 3:56 PM, Asif Iqbal wrote:
> # cat iddmi.pp
>
> class iddmi {
> # define pkg variable
> $pkg = $hardwareisa ? {
> "sparc" => "Iddmi_SunOS_sparc.pkg",
> "i386" => "Iddmi_SunOS_i386.pkg",
> default => undef,
> }
> $ftpsystem
On Jan 12, 2011 2:02 PM, "Patrick" wrote:
> On Jan 12, 2011, at 11:37 AM, wrote:
>> I have already done the setup which you mention below. I followed the
puppet-dashboard configuration guide to the tee,
>> which is why I’m surprised that this feature is not working. I have
verified that the pup
Thanks a lot. I setup and fired up httpd/passenger if I don't post
again on this thread, it means it fixed it :)
On Wed, Jan 12, 2011 at 4:39 PM, Denmat wrote:
> I get it too. It's not related to the compilation of the modules but I figure
> it's possibly related to the webrick server. I don
Hey there. Thanks for this detailed problem report, and especially
for the manifest that reproduces the problem. I will make sure we
generate a test case for this tomorrow so that we don't lose track of
it.
I can't promise a specific timeframe for fixing it, but that test will
make sure that doe
On Wed, Jan 12, 2011 at 14:33, Gary Law wrote:
> On 12 January 2011 19:08, Zman112233 wrote:
>
>> I'd like to create a banner for anyone logging into a list of servers
>> to see (i.e. Unauthorized access prohibited, you're being watched,
>> etc...).
>
> I have a "prettymotd" module. The init.pp r
On Wed, Jan 12, 2011 at 14:13, Jake - USPS wrote:
> I'm trying to setup my puppetmaster to use apache/mongrel. I've
> followed the guide at
> http://projects.puppetlabs.com/projects/1/wiki/Using_Mongrel
> for doing this.
>
> My puppetmaster starts fine. When I have a client hit the
> puppetmas
On Wed, Jan 12, 2011 at 12:41, Thomas Bellman wrote:
> Felix Frank wrote:
>
>> That's why I originally asked for pastes of configs, manifests *and*
>> filesystem listings (not to the list preferably, use pastebin services).
>
> I strongly disagree! Having things on a website instead of in
> the a
42 matches
Mail list logo
