On Mon, Sep 6, 2010 at 11:53 AM, Douglas Garstang
<[email protected]> wrote:
> On Mon, Sep 6, 2010 at 11:31 AM, James Turnbull <[email protected]> wrote:
>> Douglas Garstang wrote:
>>> 2010/9/6 Héctor Rivas Gándara <[email protected]>:
>>>> On Sat, Sep 4, 2010 at 10:44 PM, Douglas Garstang
>>>> <[email protected]> wrote:
>>>>
>>>>> Can someone tell me why, when I restart puppet on the client, I get this:
>>>>>
>>>>> Sep  4 20:41:49 [email protected] puppet-agent[26756]:
>>>>> Could not retrieve catalog from remote server: Error 400 on SERVER:
>>>>> Syntax error at ';'; expected '}' at
>>>>> /etc/puppet/modules/oracle/manifests/server/environment.pp:40 on node
>>>>> db01.pax.livegamer.com
>>>> I think that the problem is the .pp files end of line format and
>>>> enconding. It should be latin1 and (the most important) UNIX like end
>>>> of lines == "\n", not "\r\n" like in Windows files.
>>
>> I reiterate R.I.'s comment - this has nothing to do with 2.6.1.  This
>> isn't new behaviour.  There is a syntax error somewhere in your manifests.
>
> My post was not about my error. My post was questioning how it can be
> possible to receive two different errors between subsequent runs of
> the puppet client without making changes to the server side.
>
> Here's another example. The same error reported in two different
> locations between successive runs of the client, with no changes being
> made:
>
> Sep  6 18:14:20 [email protected] puppet-agent[11876]:
> Could not run Puppet configuration client: Parameter path failed: File
> paths must be fully qualified, not
> 'opt/jboss/current/server/tfel0/deploy/jmx-console.war/WEB-INF' at
> /etc/puppet/modules/jboss/manifests/server/instance.pp:163
>
> Sep  6 18:15:11 [email protected] puppet-agent[12596]:
> Could not run Puppet configuration client: Parameter path failed: File
> paths must be fully qualified, not
> 'opt/jboss/current/server/tfel0/conf/jboss.web' at
> /etc/puppet/modules/jboss/manifests/server/instance.pp:163
>
> James, I'm not even sure why you focused on the error itself. R.I
> clearly pointed out that he's seen this issue countless times. It
> makes debugging kinda challenging.

I've actually always considered this to be relatively comprehensible.

Client and server start with a clean slate.
Client requests module X
Server parses module X via autoloading, complains about parse error.
Client requests module X
Server tells you it can't find it as it's not going to reattempt
parsing unless the file changes.

Does it really impact upon debugging that much?

What do you think makes more sense? To not present the parsing error
to the client at all? Or to continuously try and parse manifests even
though the server thinks that they have a parse error?

I can't think of an answer that is obviously more correct than the
current behaviour in this situation, which is why I'm asking what you
think should happen.



>
> Doug.
>
> --
> You received this message because you are subscribed to the Google Groups 
> "Puppet Users" group.
> To post to this group, send email to [email protected].
> To unsubscribe from this group, send email to 
> [email protected].
> For more options, visit this group at 
> http://groups.google.com/group/puppet-users?hl=en.
>
>



-- 
nigel

-- 
You received this message because you are subscribed to the Google Groups 
"Puppet Users" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/puppet-users?hl=en.

Reply via email to