I made an attempt to switch to mongrel today and I''m running into a
strange
error message I can''t seem to track down:
debug: puppetmaster: Setting servertype to ''mongrel''
Server is not a class
Tracing through with the debugger shows:
[snip]
/foo/lib/gems/puppet-0.22.4/bin/puppetmasterd:248: case
Puppet[:servertype]
(rdb:1) n
/foo/lib/gems/puppet-0.22.4/bin/puppetmasterd:249: when "webrick"
(rdb:1) n
/foo/lib/gems/puppet-0.22.4/bin/puppetmasterd:253: when "mongrel":
(rdb:1) p Puppet[:servertype]
"mongrel"
(rdb:1) n
/foo/lib/gems/puppet-0.22.4/bin/puppetmasterd:254: require
''puppet/network/server/mongrel''
(rdb:1) n
/foo/lib/gems/puppet-0.22.4/lib/puppet/network/server/mongrel.rb:50: `Server
is not a class'' (TypeError)
from
/foo/lib/gems/activesupport-1.4.2/lib/active_support/dependencies.rb:495:in
`require''
from /foo/lib/gems/puppet-0.22.4/bin/puppetmasterd:254
from /foo/bin/puppetmasterd:16:in `load''
from /foo/bin/puppetmasterd:16
/foo/lib/gems/puppet-0.22.4/lib/puppet/network/server/mongrel.rb:50:class
Puppet::Network::Server
(rdb:1)
The version of puppet is 0.22.4, mongrel is 1.0.1, and ruby is 1.8.5.
Did I miss something?
Cheers,
Ryan
hi, Ryan Dooley wrote:> I made an attempt to switch to mongrel today and I''m running into a strange > error message I can''t seem to track down:[...]> The version of puppet is 0.22.4, mongrel is 1.0.1, and ruby is 1.8.5.i have ruby 1.8.2-1, puppet and mongrel in the same versions (debian etch with puppet from unstable). i get exactly the same error... fabian -- fabian linzberger -- http://lefant.net/ -- http://debienna.at/ - do yourself a favour: use and support free software!
On Jun 1, 2007, at 4:06 AM, Fabian Linzberger wrote:> hi, > > Ryan Dooley wrote: >> I made an attempt to switch to mongrel today and I''m running into >> a strange >> error message I can''t seem to track down: > > [...] > >> The version of puppet is 0.22.4, mongrel is 1.0.1, and ruby is 1.8.5. > > i have ruby 1.8.2-1, puppet and mongrel in the same versions (debian > etch with puppet from unstable). > > i get exactly the same error...I don''t think it''s the code, as plenty of people have it working right now. Do you have an older version of Puppet installed somewhere? -- I take my children everywhere, but they always find their way back home. --Robert Orben --------------------------------------------------------------------- Luke Kanies | http://reductivelabs.com | http://madstop.com
On 01/06/07, Fabian Linzberger <e@lefant.net> wrote:> i get exactly the same error...I do too. Running centos 5, with ruby-1.8.5-5.el5 and puppet 0.22.4 from dlutter''s yum repo.
On Jun 1, 2007, at 3:47 PM, ben wrote:> On 01/06/07, Fabian Linzberger <e@lefant.net> wrote: >> i get exactly the same error... > > I do too. > > Running centos 5, with ruby-1.8.5-5.el5 and puppet 0.22.4 from > dlutter''s yum repo.Hmm. So, I''ve found the problem in the code, but... why is it working for anyone else, then? It makes no sense. I''ve filed it. -- A person''s maturity consists in having found again the seriousness one had as a child, at play. --Friedrich Nietzsche --------------------------------------------------------------------- Luke Kanies | http://reductivelabs.com | http://madstop.com
On 02/06/07, Luke Kanies <luke@madstop.com> wrote:> Hmm. So, I''ve found the problem in the code, but... why is it > working for anyone else, then? It makes no sense.I came to the conclusion that nobody else had it working with that version, at least not without patching the code. Especially since once that is fixed there are other issues, such as there being no daemonize method in mongrel, which also causes it to fail to start. See Marcin Owsiany''s email of today. In #654 you comment about it being harder to test. Are you referring to it being hard to test that mongrel starts up? Do you require a unit test before you''ll accept a patch to fix it? If so, what should it test? If there are a lot of people who have mongrel working, could they please step forward, if only to re-assure us that this is tested technology ;-) G
On Jun 6, 2007, at 12:59 PM, Graham Bleach wrote:> > I came to the conclusion that nobody else had it working with that > version, at least not without patching the code. Especially since once > that is fixed there are other issues, such as there being no daemonize > method in mongrel, which also causes it to fail to start. See Marcin > Owsiany''s email of today.Yeah, you appear to be right. As mentioned, I''ve modified the Mongrel page to mention this, and I''ll get them fixed in the next release (which is coming as soon as I can get tests passing and such).> In #654 you comment about it being harder to test. Are you referring > to it being hard to test that mongrel starts up? Do you require a unit > test before you''ll accept a patch to fix it? If so, what should it > test?Well, it''s hard to test that mongrel is working properly, but I can at least write simple unit tests to make sure the class loads and responds to daemonize(). I''ll get that done and committed today, fwiw.> If there are a lot of people who have mongrel working, could they > please step forward, if only to re-assure us that this is tested > technology ;-)At least Jeff McCune and I had it working when I originally did the work, and I was under the misimpression than Marcin also had it working. I apparently did not test it again when I changed the heirarchy of the networking code. -- Sometimes I think we''re alone. Sometimes I think we''re not. In either case, the thought is staggering. --R. Buckminster Fuller --------------------------------------------------------------------- Luke Kanies | http://reductivelabs.com | http://madstop.com
On 06/06/07, Luke Kanies <luke@madstop.com> wrote:> Well, it''s hard to test that mongrel is working properly, but I can > at least write simple unit tests to make sure the class loads and > responds to daemonize(). I''ll get that done and committed today, fwiw.Thanks. I''m wondering if you think there''s any value in smoke tests. Namely, starting puppetmasterd, puppetd and performing a few simple checks. I''m considering scripting the instructions in http://www.reductivelabs.com/trac/puppet/wiki/TestingGuide as a first step. G
On Jun 6, 2007, at 3:04 PM, Graham Bleach wrote:> > Thanks. > > I''m wondering if you think there''s any value in smoke tests. Namely, > starting puppetmasterd, puppetd and performing a few simple checks. > I''m considering scripting the instructions in > http://www.reductivelabs.com/trac/puppet/wiki/TestingGuide as a first > step.I do that for all of those daemons already, just not using mongrel, because mongrel requires a proxy in front to handle the ssl. Now that I think of it, though, I can fake the ssl with ENV settings. So yes, I think there''s value in continuing with them. :) -- Normal is getting dressed in clothes that you buy for work and driving through traffic in a car that you are still paying for - in order to get to the job you need to pay for the clothes and the car, and the house you leave vacant all day so you can afford to live in it. -- Ellen DeGeneres --------------------------------------------------------------------- Luke Kanies | http://reductivelabs.com | http://madstop.com
On 06/06/07, Luke Kanies <luke@madstop.com> wrote:> > I''m wondering if you think there''s any value in smoke tests. Namely, > > starting puppetmasterd, puppetd and performing a few simple checks. > > I''m considering scripting the instructions in > > http://www.reductivelabs.com/trac/puppet/wiki/TestingGuide as a first > > step. > > I do that for all of those daemons already, just not using mongrel, > because mongrel requires a proxy in front to handle the ssl. > > Now that I think of it, though, I can fake the ssl with ENV settings. > > So yes, I think there''s value in continuing with them. :)And I think I''ve found that there''s value in reading the source before posting :) G