use different names. Yay rails.
> >
>
> All I remember now is that commit and Markus stroking his beard and
> shaking his head a lot. :)
That's funny. I remember it more as pulling on my beard, banging my
head on the table, and whimpering "Why, why, why?!"
-- M
--
> Most serialization formats can simply not deal with > 64 bit values as
> regular numbers. They may do horrible things like truncation, or use
> the max/min value if a value is too big, or for floating point
> drastically lose precision.
Eh. It's not that the serialization formats can't deal w
> I'm not very fond of the idea of hardwiring special support for
> Package resources into Puppet (nor for resources of any other specific
> type). I still like my "constraints" idea, to which Felix directed
> your attention, but if something like that is ever implemented then
> adopting it will
> But I also don't think there's fundamentally anything *wrong* with
> using whits, since they accurately model the notion of containment,
> with respect to order: classes have a "start", and a "finish", and
> there's stuff in between. It doesn't really feel like a hack to me.
> Ironically, they *
the directories.
> > because it will rarely, if ever need to look at the filesystem to
> > find a type.
Never looking at the filesystem could be a significant semantic change.
-- Markus
--
You received this message because you are subscribed to the Google Groups
"Puppet D
gt; > <mailto:nan...@gmail.com >> wrote:
>> >
>> > On Fri, Jul 12, 2013 at 8:19 AM, Brice Figureau
>> >
>> > <mailto:brice-...@daysofwonder.com >> wrote:
>> >
>> > On Fri, 2013-07-12 at 03:52 -0700, Mar
We just released an internally developed puppet-networkdevice module in the
hope that some other folks might be interested in it :).
It's currently still in an early stage but should be pretty usable for the
basic usecases.
-> https://github.com/uniak/puppet-networkdevice
## Overview
The Cisc
"This is a test" =~ /\bhis\b/i
=> nil
irb(main):005:0> "This is a test" =~ /\bthis\b/i
=> 0
(Note that 0, meaning the start of the string, counts as true in ruby,
while nil, meaning no match found, counts as false).
-- Markus
--
You received this message beca
era.yaml
is not tried at all.
I don't know how to properly debug this.
Sorry for the long delay,
Markus
> On Sat, Jun 9, 2012 at 3:28 AM, Markus Falb wrote:
>
>> Hi,
>> I tried the new puppetlabs_spec_helper and ran into the following.
>>
>> After I put hiera in
I stumbled about something similar, see
http://groups.google.com/group/puppet-dev/tree/browse_frm/thread/e47299ece604a2d2/38073019a03b487a
I have no idea how or where to fix this.
I use puppet 2.6.16
--
Kind Regards, Markus Falb
signature.asc
Description: OpenPGP digital signature
> We have a goal to foster development discussion from the community.
> Because of that, I am proposing we move the github notifications to a
> new list, puppet-commits. I realize this may have a consequence of
> reducing patch/commit discussion. This should be compensated by:
>
> 1. Still hav
;karmic-6GB-1' ], )
> ---
>
> and the type lines:
> ---
>newparam(:images) do
> desc "..."
>end
> ---
>
> Is this usual/known? Am I doing something wrong?
Puppet is trying to be too helpful and unwrapping your array for you.
Try changi
> This was committed such that all file diffs would go into the logs.
>
> Please be very vocal about announcing this and think about turning file diffs
> off as the default.
>
> I think that most users are quite happy about being able to see diffs at the
> command line but not in the system log
> I'm writing a custom insync? for a type and provider. So if you look
> at the resource it's clearly a hash:
>
> 10 member => { '192.168.1.100:3306' => {} ,
> 11'192.168.1.101:3306' => {} },
>
> I'm trying to figure out why @
> ...ah, and that is why they have trouble. They are trying to
> statically discover the list of data, then bind facts to those values
> once, rather than trusting to the cache layer in facter.
>
> They should pull all the logic into the setcode block, or a helper
> called from it, and accept pe
> When talking to KW he said he ran into thread safety issues, hence why
> he used a mutex. Has anyone seen this before? Its a concern because we
> should be fixing cause not symptom ...
Brice and I did a fair amount of thread safety cleanup in core at one
point, but I don't recall much (if any)
On Mon, 2011-08-29 at 21:18 -0700, Patrick wrote:
> Just to be clear, I think that should be:
>
> **By default** it demands permissions of 0700 or less, and ownership
> by the users who is trying to authenticate, or it will simply bypass
> the file and carry on to the next authentication mechanism
fragment = @resource[:name].split('+')[1..-1].join('+')
> + frag_file = "/var/lib/puppet/concat/fragments/#{group}/#{fragment}"
> +
> + if File.exist?(frag_file)
> +data = File.read(frag_file)
> +if data == @resource[:cont
should have received a copy of the GNU General Public License
> along with
> # this program. If not, see <http://www.gnu.org/licenses/>.
> #
> +
> +include Puppet::Util::Diff
> +
> Puppet::Type.newtype(:concat_fragment) do
>@doc = "Create a concat fr
> >
> > Huge +1
> >
>
> Because you've reviewed it, or because you think it's a good idea?
>
One convention that works (for me at least) is to use complex numbers
where the real part is for the actual code and the imaginary part is for
the idea. Thus (+1+1i) means "Oh yeah, we need this and
ot;@@methods" to "Schedule_methods" and see if
that works.
If it does, you can submit it as a patch; if it doesn't, try
"::Scedule_scales" and "::Scedule_methods" which is uglier but should
definitely fix it.
-- Markus
--
You received this message because
probable should be documented somewhere, but not on each instance
of the pattern.
-- Markus
--
You received this message because you are subscribed to the Google Groups
"Puppet Developers" group.
To post to this group, send email to puppet-dev@googlegroups.com.
To unsubscribe from t
+1
On Sun, 2011-06-05 at 09:12 +0100, R.I.Pienaar wrote:
> This allows rubygems to be used to extend or override puppet behavior.
>
> Unfortunately the features system uses the autoloader so checking for
> the rubygems feature causes a call loop. It's possible that a more
> elegant way to avoid
the ensure block (in other
words, make sure you're applying it on top of something with the fix for
#5318 (commit:68c106e3ef192) in it.
> running the specs with an added Thread.current[:gemsearch_directories] = nil
> in both cases doesnt kill my machine like before or make more tests fai
hat. Also, if you are going to follow the per-compile
pattern of module_dir (which seems reasonable) you should also clear the
thread var (see lib/puppet/configurer.rb around line 160) so that it
doesn't get stale.
-- Markus
--
You received this message because you are subscribed to th
-- as if
allow will pass anything that doesn't raise an error, even if
"auth.allowed?(request)" returns false. Should perhaps the "true" line
be removed, or am I just missing the intent?
-- Markus
--
You received this message because you are subscribed to the Google
> Thanks for the really nice investigation and explanation, I definitely
> +1 this patch (even though I'm unaffected).
Ditto on the +1; it has the side benefit on making the comment
describing how things work in node/environment.rb correct again.
-- M
--
You received this message because yo
rminate gibberish.
For this and similar reasons I like Dan's "fail at design/testing time
if we can't infer it" approach better here.
-- Markus
--
You received this message because you are subscribed to the Google Groups
"Puppet Developers" group.
To post to thi
ks like a good way to handle it & should work fine,
assuming things are all in the right places and called correctly, etc.
Shout if you get stuck & need help debugging it.
-- Markus
--
You received this message because you are subscribed to the Google Groups
"Puppet Developers"
attributes.first] if key_attributes.length == 1
].compact
...but could (on a type-by-type basis) include things like:
[:username,"@",:host,".",:domain]
or
[:protocol,":",:port]
...or whatever to build parsible titles as needed.
-- Mark
ually fixes it rather than just hiding
it by setting the name to the title so it can later set the title to the
name. I think what's really needed is to have hash2resource be smarter
about assigning a title...which may be what you're saying as well.
-- Markus
--
You received this message be
t; >dsl = setter.to_s.sub(/=$/, '')
>
> Yeah, that looks right to me. :)
You might want to intern it (since it was originally a symbol and the
proposed code leaves it as a string):
dsl = setter.to_s.sub(/=$/, '').intern
-- Markus
P.S. I don't think a monk
I happen to do that quite frequently I'm afraid.
> Sorry if it created some issues. I didn't expect (no pun intended) that
> it wasn't working (and I'm positive this used to work).
>
> > Expecting a stubbed method never wasn't working, however expecting a
> > stubbed method >= once behaved as expec
S --
I'm currrently looking at the specs for the ssh_authorized_key provider.
> There are two tests that give me a headache:
>
>it "should create the directory if it doesn't exist" do
> File.stubs(:exist?).with(@dir).returns false
> Dir.expects(:mkdir).with(@dir,0700)
> @provide
S --
I'm currrently looking at the specs for the ssh_authorized_key provider.
>> There are two tests that give me a headache:
>>
>>it "should create the directory if it doesn't exist" do
>> File.stubs(:exist?).with(@dir).returns false
>> Dir.expects(:mkdir).with(@dir,0700)
>> @
Brice --
> This is a rather complex set of structural changes. But to understand it
> you
> > really only need to grasp two key sections; the rest is cleanup.
>
> Can you explain for people like me that didn't had time to try to
> understand the patch what it brings, how (roughly) the internals
>
ct => ... )
after_application_run_command( :application_object => ... )
on_commandline_initialization(:command_line_object => ... )
on_application_initialization(:appliation_object => ... )
Paired-with: Daniel Pitman
Signed-off-by: Markus Roberts
---
lib/puppet/applicat
Signed-off-by: Markus Roberts
---
lib/puppet/application.rb |8
lib/puppet/util/command_line.rb | 20 +---
2 files changed, 13 insertions(+), 15 deletions(-)
diff --git a/lib/puppet/application.rb b/lib/puppet/application.rb
index 5e69bae..a028a15 100644
fact that some one-step edges are now two-step edges and
tests, event propagation, etc. need to reflect that.
Paired-with: Jesse Wolfe
Signed-off-by: Markus Roberts
---
Local-branch: feature/next/resource_application_order
lib/puppet/resource/catalog.rb | 64 -
lib
0, #6810,
and #6944 which may simplify or complicate their resolution.
Paired-with: Jesse Wolfe
Signed-off-by: Markus Roberts
---
Local-branch: feature/next/resource_application_order
lib/puppet/simple_graph.rb|8 +++
lib/puppet/transaction.rb
The preceeding changes left some rough edges in the Transactions (a short,
badly named method that was only used in one place and would be clearer in-
line, a return value that was carfully retained and never used, etc.) This
commit clears some of that up.
Signed-off-by: Markus Roberts
---
Local
d the property used
to cheat Demeter which has the apropriate lifetime and can be used to hold the
state information durring a traversal.
Paired-with: Jesse Wolfe
Signed-off-by: Markus Roberts
---
Local-branch: feature/next/resource_application_order
lib/puppet/transaction.rb |
The previous commit left only one meaningful value for the method parameter
of generate_additional_resources, making it a constant not a parameter. This
commit removes it.
Paired-with: Jesse Wolfe
Signed-off-by: Markus Roberts
---
Local-branch: feature/next/resource_application_order
lib
aining the original semantic.
Paired-with: Jesse Wolfe
Signed-off-by: Markus Roberts
---
Local-branch: feature/next/resource_application_order
lib/puppet/resource/catalog.rb | 42
lib/puppet/transaction.rb | 107 ++--
lib/p
aired-with: Jesse Wolfe
Signed-off-by: Markus Roberts
---
Local-branch: feature/next/resource_application_order
lib/puppet/type.rb |8 +---
lib/puppet/type/file.rb |2 --
lib/puppet/type/tidy.rb |4 +++-
3 files changed, 4 insertions(+), 10 deletions(-)
diff --git a/lib/p
This is a rather complex set of structural changes. But to understand it you
really only need to grasp two key sections; the rest is cleanup.
Cleanup commits:
The thrust of these is to localize the code that we are going to change (the
itterator over the dependency graph) and get all th
The recent AIX work added a dependency on Puppet::Parameter::Keyvalue in
the group type, but didn't add the requisite require, causing failures
under some load orders.
Signed-off-by: Markus Roberts
---
Local-branch: feature/next/resource_application_order
lib/puppet/type/group.rb |1
#x27;puppet/provider/package'
> > require 'uri'
> > +require 'pp'
>
> seems unnecessary to me and probably a debug leftover?
That's the only explanation I can think of. :)
-- Markus
--
You received this message because you are subscribed t
> We’re also going to introduce a concatenation function called “cat”.
>
Note there is already a concatenation operator in puppet (from
lib/puppet/parser/ast/leaf.rb):
class Concat < AST::Leaf
def evaluate(scope)
@value.collect { |x| x.evaluate(scope) }.collect{ |x| x ==
:undef ? ''
> +def self.title_patterns
> + [
> +# we have two title_patterns "name" and "name:protocol". We won't
> use
> +# one pattern (that will eventually set :protocol to nil) because
> we
> +# want to use a default value for :protocol. And that does only
> work
> +
D --
For some reason FileTest.exist? was returning false,
> and FileTest.directory? returns true.
> >>
> >> […]
> >>> At best, we have some leaky stubbing in a tests or some other
> confounding
> >>> factor (but nothing that should require changing lib-code); at worst,
> we
> >>> have so
L --
For some reason FileTest.exist? was returning false,
> and FileTest.directory? returns true.
>
>
> # Find this module in the modulepath.
> def path
> -environment.modulepath.collect { |path| File.join(path, name) }.find {
> |d| FileTest.exist?(d) }
> +environment.modulepath.collec
> >validate do |value|
> > raise Puppet::Error, "number has to be numeric, not #{value}"
> unless value =~ /^[0-9]+$/
> > -raise Puppet::Error, "number #{value} out of range" unless
> (0...2**16).include?(Integer(value))
>
> just out of curiosity: Will (0...2**16) generate
> I thing that recursive copies are the type that fits worst into the current
> code; I suggest we move that out and then wait and see if symlinks should
> follow or not.
>
The match with symlinks qua symlinks is pretty hideous as well. It just
looks a little less lumpy because it's been hammered
On Mon, Mar 21, 2011 at 8:23 AM, Todd Zullinger wrote:
> If xend is not running, xm list writes to stderr and facter propagates
> this to the user. Redirect stderr to /dev/null.
> ---
> lib/facter/util/xendomains.rb |2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/
> This is mostly to Brice, but I figure others are interested.
>>
>> We've been looking at what it will take to get the network device support
>> prototype as implemented by Brice into core, along with the fact that we'll
>> be producing the RC1 for statler on April 13, and I don't think it will ma
> On Fri, Mar 18, 2011 at 11:36 AM, Markus Roberts
> wrote:
> > One thought, looking through the code (I haven't given it a thorough read
> > yet) is that it may make sense to split the MS Windows provider off from
> the
> > posixs provider (though that'
On Fri, Mar 18, 2011 at 12:07 PM, Jay Flowers wrote:
> I finally got some more time to work on this. I ended up getting RubyMine
> setup and configured for me to develop and debug with.
>
> I logged a bug on this:
> https://projects.puppetlabs.com/issues/6773
>
> If someone can give me a clue I
This is mostly to Brice, but I figure others are interested.
>
> We've been looking at what it will take to get the network device support
> prototype as implemented by Brice into core, along with the fact that we'll
> be producing the RC1 for statler on April 13, and I don't think it will make
> t
One thought, looking through the code (I haven't given it a thorough read
yet) is that it may make sense to split the MS Windows provider off from the
posixs provider (though that's out of scope here, I think).
-- M
---
When in trouble or in
>> I am very new to yaml so I have to research your suggestions. It
> >> sounds like I'll need to create an object for the yaml output to map
> >> to though.
> >>
> >> I'll respond back in the thread once I figure this out.
> >
> > It looks likr pyyaml does support loading foreign types (by mappin
On Tue, Mar 15, 2011 at 3:45 PM, Corey Osman wrote:
> I am very new to yaml so I have to research your suggestions. It
> sounds like I'll need to create an object for the yaml output to map
> to though.
>
> I'll respond back in the thread once I figure this out.
>
It looks likr pyyaml does supp
Corey --
I mentioned this before in a previous post but I can't find that post.
>
> Anyways, I am using the puppet api rest interface and my python yaml
> parser doesn't like the output given. Exactly what parser are you
> guys using to encode your objects into Yaml? I don't know why my
> parser
;ve updated the ticket for this bug (
http://projects.puppetlabs.com/issues/5010) to make it clearer what's going
on. Sorry for any headaches it caused you.
-- Markus
---
When in trouble or in doubt, run in circles,
scream and shout.
icode; the routine that emits that
message explicitly permits "no" so...?
-- Markus
---
When in trouble or in doubt, run in circles,
scream and shout. -- 1920's parody of the
maritime general prudential rule
---
J --
On Sun, Mar 13, 2011 at 3:29 PM, Jay Flowers wrote:
> Markus,
> I found the getting the source page and am trying to follow it. I can
> install facter. I get the following error when trying to install puppet:
>
> C:\Projects\Puppet\puppet>ruby install.rb
> Missing ro
sion.
As for #6693, it is ringing a bell (possibly a regression which was fixed
but then reintroduced?) but I'm not finding it in a quick search. I'll
update the ticket if I do.
-- Markus
---
When in trouble or in doubt, run in cir
','local','/Volumes/foo_disk')
{
# verify that we don't try to call "mount" redundantly
@mounted.should == false
check_fstab
@mounted = true
''
}
Puppet::Util::ExecutionStub('/sbin/umount','/Volumes/foo_disk'
> - Vendor the "json_shape" library, a lightweight JSON schema validator (
> https://github.com/jes5199/json_shape).
>
> I'm fully in support of this proposal, with one caveat: The definition of
> 'vendor'. Do you mean creating a vendor directly in the puppet repo and
> adding it as a gem there,
P --
I'm not particularly proud of this code, but I don't see an alternative that
> I like better. Eliminating Puppet::Util::ExecutionStub in favor of a global
> variable seems very clumsy (not to mention that it would leave no place to
> put comments warning that it is intended for spec testing
Am I missing something, or is this:
--- /dev/null
> +++ b/lib/puppet/util/execution_stub.rb
> @@ -0,0 +1,26 @@
> +module Puppet::Util
> + class ExecutionStub
> +class << self
> + # Set a stub block that Puppet::Util.execute() should invoke instead
> + # of actually executing command
N --
What changed in 2.6.x with regard to functional settings in [environment]
> blocks?
>
> Have we documented the list of functional settings anywhere?
>
> What settings are currently not functional in an [environment] block and
> would be desirable to make functional?
>
I'm not sure I under st
So going back to the initial proposal that James made:
It occurs to me that the logical extension of a Dashboard RBAC system (or
perhaps even moving elements of the problem upstream) is for auth.conf to
recognize users or perhaps better "roles" as an authentication construct.
This seems like a ve
ight be forced to change by things we
discovered once it saw use and 2) the nested array stricture was seen as a
little awkward for users. But by this point, yeah, it's probably better to
have it documented than to have it languish in obscurity.
-- Markus
--
J --
> It occurs to me that the logical extension of a Dashboard RBAC system
> > (or perhaps even moving elements of the problem upstream) is for
> > auth.conf to recognize users or perhaps better "roles" as an
> > authentication construct.
> >
> >
> > I like. There would be some
J --
It occurs to me that the logical extension of a Dashboard RBAC system
> (or perhaps even moving elements of the problem upstream) is for
> auth.conf to recognize users or perhaps better "roles" as an
> authentication construct.
>
I like. There would be some details that should be sorted out
> The title_pattern method is used in resource.rb in the parse_title
> method. It will expect that you return an (not really intuitive) array
> of the form
>
> [ [ regex1, ARRAY ], [ regex2, ARRAY ] ]
>
> with ARRAY = [ [:namevar1, proc_that_is_called_with_namevar1], ...
> ,[:namevarN, proc ] ]
>
>
> What I mean is this:
>
> * My understanding is that the goal is to fail when any warning or higher
> logs get sent that the tests don't expect
>
> * My experience is that immediate failure with resulting stack trace is
> always easier to debug than delayed failure
>
> * Thus, my proposal, which
on, I suspect it probably raises
~3 more; at least, that's what always happened to me when we were first
exploring this space, and Dan followed a similar trajectory as I recall.
Feel free to ask them as they occur to you.
-- Markus
---
Whe
>
> > > This would terminate the operating code inside puppet at that point;
>
>> > now logging is fatal, and there is no mechanism to suppress that. If
>> > it wasn't for that property I would pretty much entirely agree.
>>
>> Huh; not sure why it's a bad thing to terminate the processing here,
> > This would terminate the operating code inside puppet at that point;
> > now logging is fatal, and there is no mechanism to suppress that. If
> > it wasn't for that property I would pretty much entirely agree.
>
> Huh; not sure why it's a bad thing to terminate the processing here, since
> th
+1
On Wed, Mar 2, 2011 at 12:47 PM, James Turnbull wrote:
> This came from the use of the @@colormap class variable.
>
> The variables has been changed to a constant.
>
> Signed-off-by: James Turnbull
> ---
> lib/puppet/util/log/destinations.rb |6 +++---
> 1 files changed, 3 insertions(+),
On Wed, Mar 2, 2011 at 12:47 PM, James Turnbull wrote:
> The kick application has the option to ping hosts. On 1.9.x this
> code was returning "Invalid next". The next in this code has been
> replaced with an exit. Testing has confirmed this doesn't change the
> behavior of puppet kick.
>
+1
The
R.I. --
>> I'm not sure that the "how it got in there" part is irrelevant (for
>> instance, I'd like if you could confirm that state.yaml shows
>> type=>absent on a node that has not been upgraded, and note the
> yes, all my nodes have it - before upgrading. and people on IRC also
> has it witho
gt; fix at the same time we start auditing. (Without the fix, it's
> > > much worse,
> > > as R.I. documented).
> >
> > all agreed. Thanks for clarifying and tightening the language Markus.
> >
> > I mis-characterized the situation because I was unde
N --
> That is reassuring.
> >
> > I think if we peel away:
> >
> > * the "always auditing" bug
>
> Yes. I believe this is fixed by Jesse's patch you reviewed:
>
>
> https://github.com/puppetlabs/puppet/commit/e3dfe41ce7da108fc345e58c7df8c1576ea951a0
>
> > * the "audits notify" surprise
>
> This i
If you were on 2.6.4 (and probably earlier) and
> if you managed a file with the source parameter and
> if the local file was absent or content differed, we wrote out:
>
>!ruby/sym type: !ruby/sym absent
!ruby/sym owner: !ruby/sym absent
>
> to state.yaml, even if you weren't auditing.
I'm having trouble framing a response, in part because I think the question
is posed at the wrong level.
Specifically, rather than talking about the "symbols" that are being written
to state.yaml we should be asking about the conditions that they signify.
The question as posed sounds as if it's a
On Tue, 2011-03-01 at 10:20 -0800, Matt Robinson wrote:
> On Tue, Mar 1, 2011 at 5:29 AM, James Turnbull wrote:
> > James Turnbull wrote:
> >>
> >> diff --git a/lib/facter/util/virtual.rb b/lib/facter/util/virtual.rb
> >> index 129448e..06b1b6d 100644
> >> --- a/lib/facter/util/virtual.rb
> >> +++
; or some such to skip the stuff I don't care about.
Not sure if that's useful info, but I figured it might be.
-- Markus
---
When in trouble or in doubt, run in circles,
scream and shout. -- 1920's parody of the
maritime general
> > > What are you updating from? Is it possibly that this is related to
> > > the update and not the audit bug...that is, if you update a node to
> > > a version with the audit bug fixed, such as the current 2.6.next
> > > (HEAD), does it restart the services?
> >
> > 2.6.4 release to 2.6.5 releas
> > What are you updating from? Is it possibly that this is related to
> > the update and not the audit bug...that is, if you update a node to
> > a version with the audit bug fixed, such as the current 2.6.next
> > (HEAD), does it restart the services?
>
> 2.6.4 release to 2.6.5 release. If I upd
R.I. --
> > Worse it seems changing auditing sends notifies:
> > >
> > > Mar 1 08:55:57 monitor2 puppet-agent[9797]:
> > > (/Stage[main]/Mcollective::Config/File[/etc/mcollective/policies/])
> > > Scheduling refresh of Service[mcollective]
> > > Mar 1 08:56:01 monitor2 puppet-agent[9797]:
> > >
I am not a DBA, but I've seen the following pattern play out many times: the
DBA says "normalize, normalize!" the developers say "normalizing is a pain,
and in this particular case it doesn't buy us much, etc." and down the road
the DBA turns out to have been right. Then the developers rediscover
B -
> + def self.listeners_of(label)
> > +synchronize {
> > + @listeners_of[label] ||= @listeners.select do |k,l|
> > +l.listen_to?(label)
> > + end
> > +}
> > + end
> >
> >
> > Does this (due to memoization) omit listeners who are added afte
B --
> 2) Would you get a more useful stack-trace without the rescue?
>
> Actually the first version didn't had the rescue, but I had some issues
> and I wanted to print the exception, it apparently escaped the garbage
> collection.
>
> Now, I thought that a straight raise would rethrow the except
B --
> + # Triggers an instrumentation
> + #
> + # Call this method around the instrumentation point
> + # Puppet::Util::Instrumentation.instrument(:my_long_computation) do
> + # ... a long computation
> + # end
> + #
> + # This will send an event to all the listeners of "my_long_
B --
Thoughts:
1) Wow.
> +def enable
> + raise "Probe already enabled" if enabled?
> + method = @method
> + label = @label
> + data = @data
> + klass.class_eval {
> +alias_method("instrumented_#{method}", method)
> +define_method(method) do |*args|
>
>> 2) you have to be careful with interpolation, in any case, as there are
> > scope changes to watch out for.
>
> Can you elaborate on this?
>
Sure. You can do this, and it works:
x = [1,2,3]
class << x
def sum_of_squares
collect { |x| x*x }.inject { |a,b| a+b }
end
end
p x.su
P & S --
The one I currently dont know how to address (besides splitting the
> current type in two fstab-only mount-only types) is #5991
>
I have to say that this one appeals to me a lot for some reason, though I'm
not as deep in the weeds as the two of you and may be missing subtleties. I
think
1 - 100 of 1155 matches
Mail list logo
