On Oct 27, 2009, at 5:01 PM, R.I.Pienaar wrote:
>
> hello,
>
> - "Luke Kanies" wrote:
>
>> On Oct 27, 2009, at 4:37 PM, Markus Roberts wrote:
>>
For instance, I know Volcane has mentioned that the reason he
overrides environments on clients is because it's the only way he
can
On Oct 27, 2009, at 4:58 PM, Nigel Kersten wrote:
>
> On Tue, Oct 27, 2009 at 4:32 PM, Luke Kanies wrote:
>>
>> On Oct 27, 2009, at 3:09 PM, Markus Roberts wrote:
>>
>>> So one thought, what if there were one (and only one) place to get
>>> the environment but the value could optionally be an ex
hello,
- "Luke Kanies" wrote:
> On Oct 27, 2009, at 4:37 PM, Markus Roberts wrote:
>
> > > For instance, I know Volcane has mentioned that the reason he
> > > overrides environments on clients is because it's the only way he
> > > can force ordering.
> >
> > As in having proto-environments
On Tue, Oct 27, 2009 at 4:32 PM, Luke Kanies wrote:
>
> On Oct 27, 2009, at 3:09 PM, Markus Roberts wrote:
>
>> So one thought, what if there were one (and only one) place to get
>> the environment but the value could optionally be an expression/
>> lambda/special token of some sort that would pu
Agreed.
On Tue, Oct 27, 2009 at 4:42 PM, Luke Kanies wrote:
>
> On Oct 27, 2009, at 4:37 PM, Markus Roberts wrote:
>
> > > For instance, I know Volcane has mentioned that the reason he
> > > overrides environments on clients is because it's the only way he
> > > can force ordering.
> >
> > As in
On Oct 27, 2009, at 4:37 PM, Markus Roberts wrote:
> > For instance, I know Volcane has mentioned that the reason he
> > overrides environments on clients is because it's the only way he
> > can force ordering.
>
> As in having proto-environments that you sequentially pass through
> on the way
> For instance, I know Volcane has mentioned that the reason he
> overrides environments on clients is because it's the only way he
> can force ordering.
As in having proto-environments that you sequentially pass through on the
way to the final environment (no skipping stages!) or am I missing the
> 1. The environment comes from the client's conf file, unless
> > overridden by the command line. Normally, this is a literal string
> > value, but it could be one of several special values such as
> > "!!FACT[fact name]" or "!!NODE_TOOL" or some such. If one of these is
> > used the appropriate
On Oct 27, 2009, at 3:09 PM, Markus Roberts wrote:
> So one thought, what if there were one (and only one) place to get
> the environment but the value could optionally be an expression/
> lambda/special token of some sort that would pull the value from
> some other (specified) location?
>
>
Perfect - +1
On Oct 27, 2009, at 3:48 PM, Eric Sorenson wrote:
>
> Adds an ext/regexp_nodes/ directory containing a regexp-based
> classifier and some example files for it.
>
> Signed-off-by: Eric Sorenson
> ---
> ext/regexp_nodes/classes/databases |2 +
> ext/regexp_nodes/classes/
- "Markus Roberts" wrote:
> > So one thought, what if there were one (and only one) place to get
> the
> > environment but the value could optionally be an
> > expression/lambda/special token of some sort that would pull the
> value
> > from some other (specified) location?
> >
> > That get
Adds an ext/regexp_nodes/ directory containing a regexp-based
classifier and some example files for it.
Signed-off-by: Eric Sorenson
---
ext/regexp_nodes/classes/databases |2 +
ext/regexp_nodes/classes/webservers |2 +
ext/regexp_nodes/parameters/environment/prod |
> > So one thought, what if there were one (and only one) place to get the
> > environment but the value could optionally be an
> > expression/lambda/special token of some sort that would pull the value
> > from some other (specified) location?
> >
> > That gets rid of the whole issue of what-trump
hey,
- "Markus Roberts" wrote:
> So one thought, what if there were one (and only one) place to get the
> environment but the value could optionally be an
> expression/lambda/special token of some sort that would pull the value
> from some other (specified) location?
>
> That gets rid of t
So one thought, what if there were one (and only one) place to get the
environment but the value could optionally be an expression/lambda/special
token of some sort that would pull the value from some other (specified)
location?
That gets rid of the whole issue of what-trumps-what, or at least sid
Can you put it all under a common directory?
Otherwise, +1.
On Oct 27, 2009, at 2:42 PM, Eric Sorenson wrote:
>
> Add a few slightly-more-than-trivial examples and point the
> classifier at these directories out of the box.
>
> Signed-off-by: Eric Sorenson
> ---
> ext/regexp_classes/databases
It matches a given node's names against lists of regexps found on
the filesystem and sets classes and parameters where it finds matches.
Also added some examples of both under their respective directories
and set up the script so it will use them as input out of the box.
Signed-off-by: Eric Soren
Added some example files to set classes and parameters via regexp.
Modified the classifier to use the example directories out of the box.
Added '= Examples' doc section with some (perhaps redundant) explanation.
--~--~-~--~~~---~--~~
You received this message beca
Add a few slightly-more-than-trivial examples and point the
classifier at these directories out of the box.
Signed-off-by: Eric Sorenson
---
ext/regexp_classes/databases |2 ++
ext/regexp_classes/webservers |2 ++
ext/regexp_parameters/environment/prod|1
> How would the command line work relative to external nodes?
>
> 1) command line
> 2) external nodes
> 3) facts
> 4) config file
> 5) server-side default
>
> ?
>
> This is, um, heinous and hideous. Do we really need that much
> flexibility in how people use environments? Is there some way
> How would the command line work relative to external nodes?
as Paul outlined, we might want to force environments, as clients
shouldn't be able to switch into a different environment and we don't
have any possiblility to restrict access to environment.
So I think it should rather be this way:
+1
On Oct 27, 2009, at 8:42 AM, Todd Zullinger wrote:
> Mattias Saou ran into a serious bug cause by 9dff71 (Use -p option to
> killproc). The puppetmaster init script lacked a pidfile variable,
> which ended up with /usr/sbin/puppetmasterd being removed.
> ---
>
> This patch is of the brown pa
On Oct 27, 2009, at 8:26 AM, Nigel Kersten wrote:
>
> On Tue, Oct 27, 2009 at 1:11 AM, Ohad Levy wrote:
>> On Tue, Oct 27, 2009 at 4:00 PM, R.I.Pienaar wrote:
>>>
>>> How was it in 0.24? I think the most sane priority order is:
>>>
>>> - external nodes
>>> - facts
>>> - config file
>>>
>> +1
>
Nice. It'd be good to see some example files, though.
On Oct 27, 2009, at 12:50 PM, Eric Sorenson wrote:
>
>
> Signed-off-by: Eric Sorenson
> ---
> ext/regexp_nodes.rb | 199 ++
> +
> 1 files changed, 199 insertions(+), 0 deletions(-)
> create m
Signed-off-by: Eric Sorenson
---
ext/regexp_nodes.rb | 199 +++
1 files changed, 199 insertions(+), 0 deletions(-)
create mode 100644 ext/regexp_nodes.rb
diff --git a/ext/regexp_nodes.rb b/ext/regexp_nodes.rb
new file mode 100644
index 000.
Markus Roberts wrote:
> Is the delete puppetmasterd bug in 0.25.1?
Unfortunately, yes. It "only" affects those using the init scripts
from conf/redhat and without using multiple ports for puppetmasterd.
So mongrel and passenger users are safe. But in the default setup,
it's broken.
I really sho
>> We'll work on getting those into the official repos soon-ish (barring
>> any other bugs like the one I introduced to delete puppetmasterd.)
>>
>
> Is the delete puppetmasterd bug in 0.25.1?
yes.
cheers pete
--~--~-~--~~~---~--~~
You received this message becau
On Tue, Oct 27, 2009 at 10:23 AM, Todd Zullinger wrote:
> James Turnbull wrote:
> > Puppet 0.25.1 - code name "zoot" - is now available. The 0.25.1
> > release is a maintenance release in the 0.25.x branch.
>
> For the moment, I've got packages for Fedora and RHEL/CentOS at
>
>http://tmz.fed
James Turnbull wrote:
> Puppet 0.25.1 - code name "zoot" - is now available. The 0.25.1
> release is a maintenance release in the 0.25.x branch.
For the moment, I've got packages for Fedora and RHEL/CentOS at
http://tmz.fedorapeople.org/repo/puppet/
We'll work on getting those into the offi
This patch works, but is deprecated in favour of a more general solution
that's in the works. Specific responses to comments below.
> -if argtitle.nil?
> > -if argtype.is_a?(Puppet::Type)
> > -self.title = argtype.title
> > -self.type = argtype.
Mattias Saou ran into a serious bug cause by 9dff71 (Use -p option to
killproc). The puppetmaster init script lacked a pidfile variable,
which ended up with /usr/sbin/puppetmasterd being removed.
---
This patch is of the brown paper bag variety. It's an embarrassing
oversight on my part and I de
On Tue, Oct 27, 2009 at 1:11 AM, Ohad Levy wrote:
> On Tue, Oct 27, 2009 at 4:00 PM, R.I.Pienaar wrote:
>>
>> How was it in 0.24? I think the most sane priority order is:
>>
>> - external nodes
>> - facts
>> - config file
>>
> +1
So this all seems related to this bug of mine:
http://projects.re
On Tue, Oct 27, 2009 at 4:00 PM, R.I.Pienaar wrote:
>
> How was it in 0.24? I think the most sane priority order is:
>
> - external nodes
> - facts
> - config file
>
> +1
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Grou
hello,
- "Luke Kanies" wrote:
> * Change the default environment on the client to a special "ask
> >> the
> server" environment. Clients that specified an environment in
> >> their
> configuration or on the command line would use that environment.
> Clients that did not
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Puppet 0.25.1 - code name "zoot" - is now available. The 0.25.1 release
is a maintenance release in the 0.25.x branch.
The release is available at:
http://reductivelabs.com/downloads/puppet/puppet-0.25.1.tar.gz
http://reductivelabs.com/downloads/ge
35 matches
Mail list logo
