This Splunk app provides custom source types and views into the status of Puppet installations that are configured to send reports, facts and metrics with the splunk_hec
, puppet_metrics_collector
and pe_status_check
Puppet modules.
You can take action on this data by connecting a Splunk installation to the PE Orchestration service via the Puppet Alert Orchestrator add-on for Splunk.
Once the application has been installed follow the steps below to configure the Puppet Report Viewer app for Splunk:
Create an Splunk HEC token for the app:
Settings
> Data Input
.HTTP Event Collector
token with a name of your choice.indexer acknowledgement
is not enabled.main
indexmain
.After configuring splunk_hec
, puppet_metrics_collector
, and pe_status_check
the Overview tab will start showing data from Puppet reports, while the PE Metrics tab will start displaying graphs related to a number of useful Puppet metrics and results of status checks.
This app includes the following custom source types:
Data from puppetlabs-splunk_hec module:
puppet:facts
puppet:summary
Data from puppetlabs-puppet_metrics_collector module:
puppet:metrics
Data from puppetlabs-pe_event_forwarding module:
puppet:activities_classifier
puppet:activities_code_manager
puppet:activities_console
puppet:activities_rbac
puppet:activity
puppet:events_summary
puppet:jobs
Data from Puppet Alert Orchestrator add-on for Splunk:
puppet:action
puppet:bolt
puppet:detailed
Data from Puppet infrastructure nodes via Splunk Forwarder:
puppet:service_logs
puppet:access_logs
Note: Access and Service logs for the following Puppet services are currently supported.
All dashboard views support using custom indexes for storing event data. The searches assume each sourcetype can be stored in it's own index. There is one top level macro (puppet_index
) which defaults to ""
.
As such, you can specify different HEC tokens for Summary Reports, Facts, and Metrics in splunk_hec
. If you configure multiple HEC tokens to utilize different indexes, change the index
value in the dashboards to reflect this.
Create five indexes named puppet_data
, puppet_summary_data
, puppet_facts_data
, puppet_metrics_data
, and puppet_detailed_data
.
Create five HEC tokens:
puppet_data
with sourcetype of automatic
and the index puppet_data
.puppet_summary
with sourcetype of puppet:summary
and the index puppet_summary_data
.puppet_facts
with sourcetype of puppet:facts
and the index of puppet_facts_data
.puppet_metrics
with sourcetype of puppet:metrics
and the index of puppet_metrics_data
.puppet_detailed
with sourcetype of puppet:detailed
and the index of puppet_detailed_data
.Configure the splunk_hec
module with the following token parameters:
splunk_hec::token
with the puppet_data
token value.splunk_hec::token_summary
with the puppet_summary
token value.splunk_hec::token_facts
with the puppet_facts
token value.splunk_hec::token_metrics
with the puppet_metrics
token value.Update the search macros to use the new values:
Settings
> Knowledge
.Search Macros
.puppet_index
and change the definition to index=puppet_data
, click save.puppet_summary_index
and change the definition to index=puppet_summary_data
, click save.puppet_facts_index
and change the definition to index=puppet_facts_data
, click save.puppet_metrics_index
and change the definition to index=puppet_metrics_data
, click save.puppet_detailed_index
and change the definition to index=puppet_detailed_data
, click save.Upon reloading the Overview tab in the Puppet Report Viewer app, and you should begin seeing data. Alternatively, perform the following search:
`puppet_all_index` sourcetype=puppet:*
Top API Calls
`puppet_logs_index` sourcetype=puppet:access_logs source=*puppetserver-access.log
| rename request as Endpoint
| chart avg(elapsed_time) as "Time (ms)" by Endpoint
| sort - "Time (ms)"
Longest Compile Time
`puppet_logs_index` sourcetype=puppet:access_logs source=*puppetserver-access.log method=POST request=/puppet/v3/catalog*
| chart max(elapsed_time) as "Time (ms)" by client
| sort - "Time (ms)"
Longest Query Time
`puppet_logs_index` sourcetype=puppet:access_logs source=*puppetdb-access.log request=/pdb/query/*
| chart max(elapsed_time) as "Time (ms)" by client
| sort - "Time (ms)"
Sync Issues
`puppet_logs_index` sourcetype=puppet:service_logs service=sync log_level=WARN OR ERROR
| chart values(message) as Message by log_level
If the Puppet Report Viewer does not appear to show any data after you have followed the configuration steps for both this app and the splunk_hec module; first check that data is being successfully sent to the Splunk server by following the troubleshooting and verification steps in the splunk_hec
documentation.
If events in the puppet:detailed
source type is not showing up in search, it means that the "Generate a Detailed Report" Alert is not configured properly with the Puppet Alert Orchestrator add-on. If this Alert is enabled, and the aforementioned add-on is configured, you can view the logs with the following Splunk search:
index=_internal sourcetype=splunkd component=sendmodalert (action="puppet_run_task_investigate" OR action="puppet_run_task" OR action="puppet_run_task_act" OR action="puppet_generate_detailed_report")
There is also a view into the Alert Actions logs under the Actions tab which will show these searches as well.
If there are no error messages, verify that the configured HEC works can be used to submit a report manually:
$ curl -k -H "Authorization: Splunk <yourHECtoken>" https://localhost:8088/services/collector/event -d '{"sourcetype": "puppet:detailed", "event": "exampletest"}'
The endpoint will respond with either a success or an error message. Follow steps in the HTTP Event Collector documentation to resolve issues with the HEC endpoint.
If there are other error messages in the logs related to PE RBAC tokens from the Puppet side, run the following command to query the PuppetDB API:
curl -k 'https://<your.puppetdb.server>:8081/pdb/query/v4/nodes' -H "X-Authentication: <token contents>"