Jira (PDB-239) After a change causing a service to not restart, it takes two puppet run for Puppetdb's events API to get the proper status

[Philippe Godin (JIRA)]

Title: Message Title

Philippe Godin commented on an issue   Re: After a change causing a service to not restart, it takes two puppet run for Puppetdb's events API to get the proper status I agree with the reporter. It seems that puppet does not consider a failure to restart as a failure. Here's the last run summary obtained from the node: [root@managed-web ~]# cat /var/lib/puppet/state/last_run_summary.yaml � time: schedule: 0.001073 last_run: 1389108067 service: 1.006371 notify: 0.006178 filebucket: 0.000199 file_line: 0.000418 total: 14.755456094162 resources: 0.000189 group: 0.001329 user: 0.005374 package: 0.135574 exec: 0.580601 anchor: 0.001642 mount: 0.001635 file: 0.861216 firewall: 0.004455 yumrepo: 0.014863 config_retrieval: 12.134339094162 resources: scheduled: 0 out_of_sync: 5 total: 201 skipped: 0 changed: 5 restarted: 0 failed: 0 failed_to_restart: 1 events: success: 5 failure: 0 total: 5 changes: total: 5 version: config: 1389052490 puppet: "3.3.1" Here's the last part of the last run report obtained from the node: resources: !ruby/object:Puppet::Util::Metric name: resources label: Resources values: - scheduled Scheduled 0 - out_of_sync "Out of sync" 5 - skipped Skipped 0 - total Total 201 - changed Changed 5 - restarted Restarted 0 - failed Failed 0 - failed_to_restart "Failed to restart" 1 changes: !ruby/object:Puppet::Util::Metric name: changes label: Changes values: - total Total 5 transaction_uuid: f915f048-c16a-4ecc-a423-0fbc20e90f07 time: 2014-01-07 10:20:39.760973 -05:00 configuration_version: 1389052490 environment: production status: changed kind: apply host: managed-web.example.org As we can see, the report status is "changed" and the metrics are reported properly. Currently, there's no way to query puppetdb to get such an event. Perhaps this issue is also linked to PDB-16. Add Comment   PuppetDB / PDB-239 After a change causing a service to not restart, it takes two puppet run for Puppetdb's events API to get the proper status When using: 'query': '["and", ["=", "certname", "mynode.localdomain"],["=", "latest-report?", "true"]]' On endpoint: /v3/events All 'event' elements returned will have their status at "success" even after a puppet run have a service ressource that failed to restart. A following puppet run (a second try) will give us the proper 'failure' status. ... This message was sent by Atlassian JIRA (v6.1.4#6159-sha1:44eaede) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at http://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/groups/opt_out.