Jeff
2009-Feb-19 16:23 UTC
[Puppet Users] One troublesome puppet hangs, an identical puppet does not
I have two servers with an identical configuration. Let''s call them 002 and 003. I''ve narrowed down my problem to the following two blocks of puppet code: file {[ "/usr/local/jboss/server", "/usr/local/ATG/ATG2007.1/home/logs", "/usr/local/ATG/ATG2007.1/home/PublishingAgent/data"]: owner => "jboss", group => "jboss", recurse => true, } file { "atg-oracle-ds.xml": path => "/usr/local/jboss/server/default/deploy/atg-oracle- ds.xml", mode => 0644, ensure => present, content => template("/var/puppet/modules/atg/files/$fqdn/atg-oracle- ds.xml.erb"), require => Package["ATG"] } Each time they get to the file atg-oracle-ds.xml I get a warning that says, debug: /Class[main]/Node[lsnas003]/Class[atg::config]/File[/usr/local/ jboss/server/default/deploy]: Not managing more explicit file /usr/ local/jboss/server/default/deploy/atg-oracle-ds.xml The difference between the two servers is pretty significant: 002 pauses and continues its run but 003 hangs. If I comment out the first block, i,e., #file {[ # "/usr/local/jboss/server", # "/usr/local/ATG/ATG2007.1/home/logs", # "/usr/local/ATG/ATG2007.1/home/PublishingAgent/data"]: # owner => "jboss", # group => "jboss", # recurse => true, #} then 003 works just fine. I don''t know if this is helpful, but I''ll pass it along. When 003 is hung, I stop it with ctrl-c. When that happens, it shows me where it caught the signal: notice: Caught INT; shutting down debug: Signal caught here: debug: /usr/lib/site_ruby/1.8/puppet/metatype/instances.rb:199:in `call'' debug: /usr/lib/site_ruby/1.8/puppet/metatype/instances.rb:199:in `each'' debug: /usr/lib/site_ruby/1.8/puppet/metatype/instances.rb:198:in `each'' debug: /usr/lib/site_ruby/1.8/puppet/external/gratr/adjacency_graph.rb: 110:in `include?'' debug: /usr/lib/site_ruby/1.8/puppet/external/gratr/adjacency_graph.rb: 110:in `add_edge!'' debug: /usr/lib/site_ruby/1.8/puppet/pgraph.rb:19:in `add_edge!'' debug: /usr/lib/site_ruby/1.8/puppet/transaction.rb:177:in `copy_relationships'' debug: /usr/lib/site_ruby/1.8/puppet/transaction.rb:170:in `each'' debug: /usr/lib/site_ruby/1.8/puppet/transaction.rb:170:in `copy_relationships'' debug: /usr/lib/site_ruby/1.8/puppet/transaction.rb:259:in `eval_resource'' debug: /usr/lib/site_ruby/1.8/puppet/transaction.rb:251:in `eval_resource'' debug: /usr/lib/site_ruby/1.8/puppet/transaction.rb:250:in `each'' debug: /usr/lib/site_ruby/1.8/puppet/transaction.rb:250:in `eval_resource'' debug: /usr/lib/site_ruby/1.8/puppet/transaction.rb:316:in `evaluate'' debug: /usr/lib/site_ruby/1.8/puppet/util.rb:434:in `thinmark'' Any ideas? I''m almost positive these two servers are identical from a hardware/OS/puppet configuration perspective. --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Puppet Users" group. To post to this group, send email to puppet-users@googlegroups.com To unsubscribe from this group, send email to puppet-users+unsubscribe@googlegroups.com For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en -~----------~----~----~----~------~----~------~--~---
Brice Figureau
2009-Feb-20 08:27 UTC
[Puppet Users] Re: One troublesome puppet hangs, an identical puppet does not
On Thu, 2009-02-19 at 08:23 -0800, Jeff wrote:> I have two servers with an identical configuration. Let''s call them > 002 and 003. I''ve narrowed down my problem to the following two blocks > of puppet code: > > file {[ > "/usr/local/jboss/server", > "/usr/local/ATG/ATG2007.1/home/logs", > "/usr/local/ATG/ATG2007.1/home/PublishingAgent/data"]: > owner => "jboss", > group => "jboss", > recurse => true, > } > > file { "atg-oracle-ds.xml": > path => "/usr/local/jboss/server/default/deploy/atg-oracle- > ds.xml", > mode => 0644, > ensure => present, > content => template("/var/puppet/modules/atg/files/$fqdn/atg-oracle- > ds.xml.erb"), > require => Package["ATG"] > } > > Each time they get to the file atg-oracle-ds.xml I get a warning that > says, > > debug: /Class[main]/Node[lsnas003]/Class[atg::config]/File[/usr/local/ > jboss/server/default/deploy]: Not managing more explicit file /usr/ > local/jboss/server/default/deploy/atg-oracle-ds.xml > > The difference between the two servers is pretty significant: 002 > pauses and continues its run but 003 hangs.This is a pure guess: is there a lot of files under those directories in 002 that are not in 003: "/usr/local/jboss/server", "/usr/local/ATG/ATG2007.1/home/logs", "/usr/local/ATG/ATG2007.1/home/PublishingAgent/data" I''ve noticed this is a file with recurse, if you happen to have checksum globally enabled puppetd will compute (and cache) the checksum for each file beneath those 3 paths. If there are _lots_ of files that can take time. If you run puppetd with --debug on 002, maybe you''ll see the root cause of your issue. -- Brice Figureau My Blog: http://www.masterzen.fr/ --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Puppet Users" group. To post to this group, send email to puppet-users@googlegroups.com To unsubscribe from this group, send email to puppet-users+unsubscribe@googlegroups.com For more options, visit this group at http://groups.google.com/group/puppet-users?hl=en -~----------~----~----~----~------~----~------~--~---