Hi all, I''ve got a Puppet 2.6.8 setup on RHEL 5.6 with nodes.pp containing a simple import "nodes/*.pp" and per-node configs in nodes/<hostname>.pp Occasionally puppet master misses new nodes'' files, simply ignores them and the subsequent puppet agent run on the node fails with "Could not find default node or by name with ...". All that is needed in such a case is "/etc/init.d/puppetmaster restart" but it''s clearly not an optimal "fix". It doesn''t happen all the time, in most cases the new node''s file is correctly recognised, but in the last case the file was sitting there for 4 days unnoticed and only after puppetmaster restart it was read in. Any idea what''s causing it? Is there anything we can do to prevent this from happening again and again? Thanks! Michal -- 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.
Hi, Puppetmasterd uses the webrick web server which is meant really for development only. Once you get over a number of nodes you see these issue. Check the puppetlabs docs on scaling for your options. The other issue you are seeing is due to caching. I know this was a recently looked into but I can''t remember if it was addressed and in which version (most likely 2.7.x). You can query the project page for current tickets. Cheers, Den On 21/11/2011, at 11:08, Michal Ludvig <mludvig@logix.net.nz> wrote:> Hi all, > > I''ve got a Puppet 2.6.8 setup on RHEL 5.6 with nodes.pp containing a simple import "nodes/*.pp" and per-node configs in nodes/<hostname>.pp > > Occasionally puppet master misses new nodes'' files, simply ignores them and the subsequent puppet agent run on the node fails with "Could not find default node or by name with ...". > > All that is needed in such a case is "/etc/init.d/puppetmaster restart" but it''s clearly not an optimal "fix". It doesn''t happen all the time, in most cases the new node''s file is correctly recognised, but in the last case the file was sitting there for 4 days unnoticed and only after puppetmaster restart it was read in. > > Any idea what''s causing it? Is there anything we can do to prevent this from happening again and again? > > Thanks! > > Michal > > > > -- > 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. >-- 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.
On 11/21/2011 04:46 PM, Denmat wrote:> Hi, > > Puppetmasterd uses the webrick web server which is meant really for development only. Once you get over a number of nodes you see these issue. Check the puppetlabs docs on scaling for your options.As I''ve got only 12 nodes in this project I don''t think I ran into a scaling issue yet..> The other issue you are seeing is due to caching. I know this was a recently looked into but I can''t remember if it was addressed and in which version (most likely 2.7.x). You can query the project page for current tickets. >I can''t find anything relevant among the recent ca 2 months of tickets. Any more specific pointers plz? Thanks Michal> Cheers, > Den > > On 21/11/2011, at 11:08, Michal Ludvig<mludvig@logix.net.nz> wrote: > >> Hi all, >> >> I''ve got a Puppet 2.6.8 setup on RHEL 5.6 with nodes.pp containing a simple import "nodes/*.pp" and per-node configs in nodes/<hostname>.pp >> >> Occasionally puppet master misses new nodes'' files, simply ignores them and the subsequent puppet agent run on the node fails with "Could not find default node or by name with ...". >> >> All that is needed in such a case is "/etc/init.d/puppetmaster restart" but it''s clearly not an optimal "fix". It doesn''t happen all the time, in most cases the new node''s file is correctly recognised, but in the last case the file was sitting there for 4 days unnoticed and only after puppetmaster restart it was read in. >> >> Any idea what''s causing it? Is there anything we can do to prevent this from happening again and again? >> >> Thanks! >> >> Michal >> >> >> >> -- >> 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. >>-- 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.
On 21/11/2011, at 15:20, Michal Ludvig <mludvig@logix.net.nz> wrote:> On 11/21/2011 04:46 PM, Denmat wrote: >> Hi, >> >> Puppetmasterd uses the webrick web server which is meant really for development only. Once you get over a number of nodes you see these issue. Check the puppetlabs docs on scaling for your options. > > As I''ve got only 12 nodes in this project I don''t think I ran into a scaling issue yet..Yeah, 12 isn''t big but the webrick isnt built to handle many (single threaded I believe), especially if they are reporting in at the same time. I only had a handful before setting up passenger/apache.>> The other issue you are seeing is due to caching. I know this was a recently looked into but I can''t remember if it was addressed and in which version (most likely 2.7.x). You can query the project page for current tickets. >> > > I can''t find anything relevant among the recent ca 2 months of tickets. Any more specific pointers plz?http://projects.puppetlabs.com/issues/8750> > Thanks > > Michal > >> Cheers, >> Den >> >> On 21/11/2011, at 11:08, Michal Ludvig<mludvig@logix.net.nz> wrote: >> >>> Hi all, >>> >>> I''ve got a Puppet 2.6.8 setup on RHEL 5.6 with nodes.pp containing a simple import "nodes/*.pp" and per-node configs in nodes/<hostname>.pp >>> >>> Occasionally puppet master misses new nodes'' files, simply ignores them and the subsequent puppet agent run on the node fails with "Could not find default node or by name with ...". >>> >>> All that is needed in such a case is "/etc/init.d/puppetmaster restart" but it''s clearly not an optimal "fix". It doesn''t happen all the time, in most cases the new node''s file is correctly recognised, but in the last case the file was sitting there for 4 days unnoticed and only after puppetmaster restart it was read in. >>> >>> Any idea what''s causing it? Is there anything we can do to prevent this from happening again and again? >>> >>> Thanks! >>> >>> Michal >>> >>> >>> >>> -- >>> 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. >>> > > -- > 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. >-- 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.
On 2011-11-20 16:08 , Michal Ludvig wrote:> Hi all, > > I''ve got a Puppet 2.6.8 setup on RHEL 5.6 with nodes.pp containing a > simple import "nodes/*.pp" and per-node configs in nodes/<hostname>.pp > > Occasionally puppet master misses new nodes'' files, simply ignores them > and the subsequent puppet agent run on the node fails with "Could not > find default node or by name with ...". > > All that is needed in such a case is "/etc/init.d/puppetmaster restart" > but it''s clearly not an optimal "fix". It doesn''t happen all the time, > in most cases the new node''s file is correctly recognised, but in the > last case the file was sitting there for 4 days unnoticed and only after > puppetmaster restart it was read in. > > Any idea what''s causing it? Is there anything we can do to prevent this > from happening again and again? > > Thanks! > > Michal > > >Pretty sure you''re just running into the "normal" behavior of the caching that puppet does: Since puppet caches successfully compiled files until it notices that they''ve changed, it won''t go back to re-compile the nodes.pp (and pick up the new nodes/<hostname>.pp files) until nodes.pp itself needs a recompile. -- Jacob Helwig http://about.me/jhelwig