Skip to content

Latest commit

 

History

History
299 lines (237 loc) · 11.4 KB

reports.markdown

File metadata and controls

299 lines (237 loc) · 11.4 KB
Raw
title layout canonical
PuppetDB 3.1 » API » v4 » Querying Reports
default
/puppetdb/latest/api/query/v4/reports.html

Puppet agent nodes submit reports after their runs, and the puppet master forwards these to PuppetDB. Each report includes:

  • Some data about the entire run
  • Some metadata about the report
  • Many events, describing what happened during the run

Once this information is stored in PuppetDB, it can be queried in various ways.

  • You can query data about the run and report metadata by making an HTTP request to the /reports endpoint.
  • You can query data about individual events by making an HTTP request to the /events endpoint.
  • You can query summaries of event data by making an HTTP request to the /event-counts or aggregate-event-counts endpoints.

/pdb/query/v4/reports

URL Parameters

  • query: Optional. A JSON array of query predicates, in prefix notation (["<OPERATOR>", "<FIELD>", "<VALUE>"]). See the sections below for the supported operators and fields. For general info about queries, see the page on query structure.

    If the query parameter is absent, PuppetDB will return all reports.

Query Operators

See the Operators page

Query Fields

The below fields are allowed as filter criteria and are returned in all responses.

  • certname (string): the name of the node that the report was received from.

  • hash (string): the id of the report; these ids can be acquired via event queries (see the /events endpoint).

  • environment (string): the environment assigned to the node that submitted the report.

  • status (string): the status associated to report's node. Possible values for this field come from Puppet's report status, which can be found here.

  • noop (boolean): a flag indicating whether the report was produced by a noop run.

  • puppet_version (string): the version of Puppet that generated the report.

  • report_format (number): the version number of the report format that Puppet used to generate the original report data.

  • configuration_version (string): an identifier string that Puppet uses to match a specific catalog for a node to a specific Puppet run.

  • start_time (timestamp): is the time on the agent at which the Puppet run began. Timestamps are always ISO-8601 compatible date/time strings.

  • end_time (timestamp): is the time on the agent at which the Puppet run ended. Timestamps are always ISO-8601 compatible date/time strings.

  • producer_timestamp (timestamp): is the time of catalog submission from the master to PuppetDB, according to the clock on the master. Timestamps are always ISO-8601 compatible date/time strings.

  • receive_time (timestamp): is the time at which PuppetDB received the report. Timestamps are always ISO-8601 compatible date/time strings.

  • transaction_uuid (string): string used to identify a Puppet run.

  • latest_report? (boolean): return only reports associated with the most recent puppet run for each node. NOTE: this field does not appear in the response.

Response format

The response is a JSON array of report summaries for all event reports that matched the input parameters. The array is unsorted. The top-level response is of the form:

{
  "hash": <sha1 of stored report command payload>,
  "puppet_version": <report puppet version>,
  "receive_time": <time of report reception by PDB>,
  "report_format": <report wireformat version>,
  "start_time": <start of run timestamp>,
  "end_time": <end of run timestamp>,
  "transaction_uuid": <string to identify puppet run>,
  "status": <status of node after report's associated puppet run>,
  "noop": <boolean flag indicating noop run>,
  "environment": <report environment>,
  "configuration_version": <catalog identifier>,
  "certname": <node name>,
  "resource_events": <expanded resource events>,
  "metrics" : <expanded metrics>,
  "logs" : <expanded logs>
}

Note: Expansion and the data field

For the following data structures the data field is only expanded for users running PostgreSQL. Expansion is not supported on HSQLDB, instead you must use the href field to construct a secondary query to retrieve that information.

The <expanded resource events> object is of the following form:

{
  "href": <url>,
  "data": [ {
    "status": <status of event (`success`, `failure`, `noop`, or `skipped`)>,
    "timestamp": <timestamp (from agent) at which event occurred>,
    "resource_type": <type of resource event occurred on>,
    "resource_title": <title of resource event occurred on>,
    "property": <property/parameter of resource on which event occurred>,
    "new_value": <new value for resource property>,
    "old_value": <old value of resource property>,
    "message": <description of what happened during event>,
    "file": <manifest file containing resource definition>,
    "line": <line in manifest file on which resource is defined>,
    "containment_path": <containment heirarchy of resource within catalog>
  } ... ]
}

The <expanded metrics> object is as follows:

{
  "href": <url>,
  "data": [ {
    "category" : <category of metric ("resources", "time", "changes", or "events")>,
    "name" : <name of the metric>,
    "value" : <value of the metric (double precision)>
  } ... ]
}

The <expanded logs> object returns all a single log line per data entry as follows:

{
  "href": <url>,
  "data": [ {
    "file" : <file of resource declaration>,
    "line" : <line of resource declaration>,
    "level" : <log level>,
    "message" : <log message>,
    "source" : <log source>,
    "tags" : [<resource tag>],
    "time" : <log line timestamp>
 } ... ]

}

File and line may each be null if the log does not concern a resource.

Note on fields that allow NULL values

In the resource_event schema above, containment_path, new_value, old_value, property, file, line, status, and message may all be null.

Note on querying resource events, metrics, and logs

The reports endpoint does not support querying on the value of resource_events, logs, or metrics. For resource_events the same information can be accessed by querying the events endpoint for events with field report equal to a given report's hash. Making metrics and logs queryable may be the target of future work.

Examples

You can use curl to query information about reports like so:

Query for all reports:

curl -G http://localhost:8080/pdb/query/v4/reports

[ {
  "receive_time" : "2015-02-19T16:23:11.034Z",
  "hash" : "32c821673e647b0650717db467abc51d9949fd9a",
  "transaction_uuid" : "9a7070e9-840f-446d-b756-6f19bf2e2efc",
  "puppet_version" : "3.7.4",
  "noop" : false,
  "report_format" : 4,
  "start_time" : "2015-02-19T16:23:09.810Z",
  "end_time" : "2015-02-19T16:23:10.287Z",
  "resource_events" : {
    "href": "/pdb/query/v4/reports/32c821673e647b0650717db467abc51d9949fd9a/events",
    "data": [ {
      "new_value" : "hi world",
      "property" : "message",
      "file" : "/home/wyatt/.puppet/manifests/site.pp",
      "old_value" : "absent",
      "line" : 7,
      "resource_type" : "Notify",
      "status" : "success",
      "resource_title" : "hiloo",
      "timestamp" : "2015-02-19T16:23:10.768Z",
      "containment_path" : [ "Stage[main]", "Main", "Notify[hiloo]" ],
      "message" : "defined 'message' as 'hi world'"
    }, {
      "new_value" : "hi world",
      "property" : "message",
      "file" : "/home/wyatt/.puppet/manifests/site.pp",
      "old_value" : "absent",
      "line" : 3,
      "resource_type" : "Notify",
      "status" : "success",
      "resource_title" : "hi",
      "timestamp" : "2015-02-19T16:23:10.767Z",
      "containment_path" : [ "Stage[main]", "Main", "Notify[hi]" ],
      "message" : "defined 'message' as 'hi world'"
    } ]
  },
  "status" : "changed",
  "configuration_version" : "1424362990",
  "environment" : "production",
  "certname" : "desktop.localdomain",
  "metrics" : {
    "href": "/pdb/query/v4/reports/32c821673e647b0650717db467abc51d9949fd9a/metrics",
    "data": [ {
      "category" : "resources",
      "name" : "changed",
      "value" : 2
    }, {
      "category" : "resources",
      "name" : "failed",
      "value" : 0
    },
    ...
    {
      "category" : "events",
      "name" : "success",
      "value" : 2
    }, {
      "category" : "events",
      "name" : "total",
      "value" : 2
    } ]
  },
  "logs" : {
    "href": "/pdb/query/v4/reports/32c821673e647b0650717db467abc51d9949fd9a/logs",
    "data": [ {
      "file" : null,
      "line" : null,
      "level" : "info",
      "message" : "Caching catalog for mbp.local",
      "source" : "//mbp.local/Puppet",
      "tags" : [ "info" ],
      "time" : "2015-02-26T16:27:48.416642000-08:00"
    },
    ...
    {
      "file" : null,
      "line" : null,
      "level" : "notice",
      "message" : "Finished catalog run in 0.01 seconds",
      "source" : "//mbp.local/Puppet",
      "tags" : [ "notice" ],
      "time" : "2015-02-26T16:27:48.483317000-08:00"
    } ]
  }
} ]

Get counts of reports by grouped by status

curl -X GET http://localhost:8080/pdb/query/v4/reports \
  -d 'query=["extract",[["function","count"], "status"],
                        ["~","certname",""],
                        ["group_by", "status"]]'

[ {
  "status" : "failed",
  "count" : 10
}, {
  "status" : "changed",
  "count" : 72
}, {
  "status" : "unchanged",
  "count" : 20
} ]

/pdb/query/v4/reports/<HASH>/events

This will return all events for a particular report, designated by its unique hash.

This is a shortcut to the /events endpoint. It behaves the same as a call to /events with a query string of ["=", "report", "<HASH>"].

/pdb/query/v4/reports/<HASH>/metrics

This will return all metrics for a particular report, designated by its unique hash. This endpoint does not currently support querying or paging.

/pdb/query/v4/reports/<HASH>/logs

This will return all logs for a particular report, designated by its unique hash. This endpoint does not currently support querying or paging.

URL Parameters / Query Operators / Query Fields / Response Format

This route is an extension of the events endpoint. It uses the exact same parameters, operators, fields, and response format.

If you provide a query parameter, it will specify additional criteria, which will be used to return a subset of the information normally returned by this route.

Paging

This query endpoint supports paged results via the common PuppetDB paging URL parameters. For more information, please see the documentation on paging.