Permalink
Fetching contributors…
Cannot retrieve contributors at this time
512 lines (410 sloc) 18.4 KB

Tasks

A task describes the how, what, and when to do for a Snap job.

Skip to the TL;DR example here.

Task States

A task can be in the following states:

  • running: a running task
  • stopped: a task that is not running
  • disabled: a task in a state not allowed to start. This happens when the task produces consecutive errors. A disabled task must be re-enabled before it can be started again.
  • ended: a task for which the schedule is ended. It happens for schedule with defined stop_timestamp or with specified the count of runs. An ended task is resumable if the schedule is still valid.

statediagram

How To Command
Create task snaptel task create [command options] [arguments...]
Find more details here
List snaptel task list
Start task snaptel task start <task_id>
Stop task snaptel task stop <task_id>
Remove task snaptel task remove <task_id>
Export task snaptel task export <task_id>
Watch task snaptel task watch <task_id>
Enable task snaptel task enable <task_id>

Task Manifest

A task is described in a task manifest, which can be either JSON or YAML1. The manifest is divided into two parts: Header and Workflow.

The Header

---
  version: 1
  schedule:
    type: "simple"
    interval: "1s"
  max-failures: 10

Version

The header contains a version, used to differentiate between versions of the task manifest parser. Right now, there is only one version: 1.

Schedule

The schedule describes the schedule type and interval for running the task. At the time of this writing, Snap has three schedules:

Snap is designed in a way where custom schedulers can easily be dropped in. If a custom schedule is used, it may require more key/value pairs in the schedule section of the manifest.

Simple Schedule
Key Type Description
interval(*) string An interval specifies the time duration between each scheduled execution; It must be greater than 0.
count uint A count determines the number of expected scheduled executions at interval seconds apart. Defaults to 0 what means no limit. Set the count to 1 if you expect a single run task.

(*) is required

  • simple "run forever" schedule:
	"version": 1,
  "schedule": {
  	"type": "simple",
  	"interval": "1s"
  },
  "max-failures": 10,
  • simple "run X times" schedule:
  "version": 1,
  "schedule": {
  	"type": "simple",
  	"interval": "1s",
  	"count": 1
  },
  "max-failures": 1,
Windowed Schedule

The windowed schedule adds a start and/or stop time for the task.

Key Type Description
interval(*) string An interval specifies the time duration between each scheduled execution; It must be greater than 0.
start_timestamp(1) string A start time for the task schedule. If not determined, the schedule will start immediately.
stop_timestamp(1) string A stop time for the task schedule. If not determined, the schedule will be running all the time until the stop command is not called.
count uint A count determines the number of expected scheduled executions at interval seconds apart. Defaults to 0 what means no limit. Set the count to 1 if you expect a single run task.

(*) is required

(1) the time must be given as a quoted string in RFC 3339 format with specific time zone offset

Notice: Specifying both the stop_timestamp and the count is not allowed. In such case, you receive a warning that the value of the count field will be ignored.

  • a regular window with determined both start and stop time:
  "version": 1,
  "schedule": {
  	"type": "windowed",
  	"interval": "1s",
  	"start_timestamp": "2016-10-27T16:00:00+01:00",
  	"stop_timestamp": "2016-10-28T16:30:00+01:00"
      },
  "max-failures": 10,
  • start schedule on start_timestamp and "run forever":
    (a window with determined only stop time)
  "version": 1,
  "schedule": {
  	"type": "windowed",
  	"interval": "1s",
  	"start_timestamp": "2016-10-27T16:00:00+01:00"
      },
  "max-failures": 10,
  • start schedule immediately and finish on stop time:
    (a window with determined only start time)
  "version": 1,
  "schedule": {
  	"type": "windowed",
  	"interval": "1s",
  	"stop_timestamp": "2016-10-28T16:30:00+01:00"
      },
  "max-failures": 10,
  • start schedule on start time and run "X times":
    (a window with determined start time and count)
  "version": 1,
  "schedule": {
  	"type": "windowed",
  	"interval": "1s",
  	"start_timestamp": "2016-10-27T16:00:00+01:00",
  	"count": 1
      },
  "max-failures": 1,
Cron Schedule

The cron schedule supports cron-like entries in interval field. More on cron expressions can be found here: https://godoc.org/github.com/robfig/cron

Key Type Description
interval(*) string An interval specifies the time duration between each scheduled execution in cron-like entries. More on cron expressions can be found here: https://godoc.org/github.com/robfig/cron.

(*) is required

  • schedule task every hour on the half hour:
   "version": 1,
   "schedule": {
       "type": "cron",
       "interval" : "0 30 * * * *"
   },
   "max-failures": 10,
Streaming Schedule
   ---
  version: 1
  schedule:
    type: "streaming"

The streaming schedule doesn't support fields such as interval and count. If those fields are provided as part of the schedule, they will simply be skipped. For more details on streaming, visit STREAMING.md

Max-Failures

By default, Snap will disable a task if there are 10 consecutive errors from any plugins within the workflow. The configuration can be changed by specifying the number of failures value in the task header. If the max-failures value is -1, Snap will not disable a task with consecutive failure. Instead, Snap will sleep for 1 second for every 10 consecutive failures and retry again.

If you intend to run tasks with max-failures: -1, please also configure max_plugin_restarts: -1 in snap daemon control configuration section.

For more on tasks, visit SNAPTEL.md.

The Workflow

---
  collect:
    metrics:
      /intel/mock/foo: {}
      /intel/mock/bar: {}
      /intel/mock/*/baz: {}
    config:
      /intel/mock:
        user: "root"
        password: "secret"
    tags:
      /intel/mock:
        experiment: "experiment 11"
      /intel/mock/bar:
        os: "linux"
    process:
      -
        plugin_name: "passthru"
        publish:
          -
            plugin_name: "file"
            config:
              file: "/tmp/published"

The workflow is a DAG which describes the how and what of a task. It is always rooted by a collect, and then contains any number of processes and publishes.

Remote Targets

Process and Publish nodes in the workflow can also target remote Snap nodes via the 'target' key. The purpose of this is to allow offloading of resource intensive workflow steps from the node where data collection is occurring. Modifying the example above we have:

---
  collect:
    metrics:
      /intel/mock/foo: {}
      /intel/mock/bar: {}
      /intel/mock/*/baz: {}
    config:
      /intel/mock:
        user: "root"
        password: "secret"
    tags:
      /intel/mock:
        experiment: "experiment 11"
      /intel/mock/bar:
        os: "linux"
    process:
      -
        plugin_name: "passthru"
        target: "127.0.0.1:8082"
        publish:
          -
            plugin_name: "file"
            target: "127.0.0.1:8082"
            config:
              file: "/tmp/published"

If a target is specified for a step in the workflow, that step will be executed on the remote instance specified by the ip:port target. Each node in the workflow is evaluated independently so a workflow can have any, all, or none of its steps being done remotely (if target key is omitted, that step defaults to local). The ip and port target are the ip and port that has a running control-grpc server. These can be specified to snapteld via the control-listen-addr and control-listen-port flags. The default is the same ip as the Snap rest-api and port 8082.

An example json task that uses remote targets:

{
  "version": 1,
  "schedule": {
    "type": "simple",
    "interval": "1s"
  },
  "max-failures": 10,
  "workflow": {
    "collect": {
      "metrics": {
        "/intel/mock/foo": {},
        "/intel/mock/bar": {},
        "/intel/mock/*/baz": {}
      },
      "config": {
        "/intel/mock": {
          "user": "root",
          "password": "secret"
        }
      },
      "process": [
        {
          "plugin_name": "passthru",
          "target": "127.0.0.1:9999",
          "process": null,
          "publish": [
            {
              "plugin_name": "file",
              "target": "127.0.0.1:9992",
              "config": {
                "file": "/tmp/snap_published_mock_file.log"
              }
            }
          ]
        }
      ]
    }
  }
}

More information about the architecture behind this can be found here.

collect

The collect section describes which metrics, indicated by namespaces, are requested to be collected.

Elements of namespace are separated by namespace separators which can be set in the task manifest as different characters, with some limitations specific for format of the task manifest. The first character in the namespace defines the namespace separator.

It is not recommended to use following characters in the task manifest as the namespace separators:

  • for YAML: |,#, $, >,*, ,,[, ],{,},!,", `,%,@
  • for JSON: \, $,", *

Some of mentioned characters may work as namespace separators but the namespace must be in double quotes (i.e."|intel|mock|foo" for YAML) or special characters must be escaped.

Metrics can be enumerated using:

a) concrete namespace

Declaring a metric's name exactly as it appears in the metric catalog (see snaptel metric list).

Metrics requested in task manifest Collected metrics
/intel/mock/foo /intel/mock/foo
/intel/mock/bar /intel/mock/bar
/intel/mock/*/baz
(dynamic metric)
/intel/mock/host0/baz
/intel/mock/host1/baz
/intel/mock/host2/baz
/intel/mock/host3/baz
/intel/mock/host4/baz
/intel/mock/host5/baz
/intel/mock/host6/baz
/intel/mock/host7/baz
/intel/mock/host8/baz
/intel/mock/host9/baz

(collect metrics for all instances of the dynamic metric)

b) specified instance of dynamic metrics

Specifying a dynamic metric refers to providing a value in place of the dynamic element in the namespace (e.g. hostname, cgroup id, etc.). It's important to remember that dynamic elements are represented by an asterisk when presented in the metric catalog. When a task manifest contains a specific instance of a dynamic metric only that instance will be collected. If the value does not exist the task will error.

Metrics requested in task manifest Collected metrics
/intel/mock/host0/baz

(specific instance of "/intel/mock/*/baz")
/intel/mock/host0/baz

(only this one metric will be collected)

c) dynamic query

Dynamic queries are those that contain:

  • wildcards * - that matches with any value in the metric namespace or, if the wildcard is in the end, with all metrics with the given prefix
  • and/or tuples of values (x;y;z) - that matches with all items separated by semicolon and works like logical and, so it gives an error if even one of these items cannot be collected
Metrics requested in task manifest Collected metrics
/intel/mock/* /intel/mock/foo
/intel/mock/bar
/intel/mock/host0/baz
/intel/mock/host1/baz
/intel/mock/host2/baz
/intel/mock/host3/baz
/intel/mock/host4/baz
/intel/mock/host5/baz
/intel/mock/host6/baz
/intel/mock/host7/baz
/intel/mock/host8/baz
/intel/mock/host9/baz

(collect all metrics with prefix "/intel/mock/")
/intel/mock/(foo;bar) /intel/mock/foo
/intel/mock/bar
/intel/mock/(host0;host1;host2)/baz /intel/mock/host0/baz
/intel/mock/host1/baz
/intel/mock/host2/baz

The namespaces are keys to another nested object which may contain a specific version of a plugin, e.g.:

---
/foo/bar/baz:
  version: 4

If a version is not given, Snap will select the latest for you.

The config section describes configuration data for metrics. Since metric namespaces form a tree, config can be described at a branch, and all leaves of that branch will receive the given config. For example, say a task is going to collect /intel/perf/foo, /intel/perf/bar, and /intel/perf/baz, all of which require a username and password to collect. That config could be described like so:

---
metrics:
  /intel/perf/foo: {}
  /intel/perf/bar: {}
  /intel/perf/baz: {}
config:
  /intel/perf:
    username: jerr
    password: j3rr

Applying the config at /intel/perf means that all leaves of /intel/perf (/intel/perf/foo, /intel/perf/bar, and /intel/perf/baz in this case) will receive the config.

The tag section describes additional meta data for metrics. Similar to config, tags can also be described at a branch, and all leaves of that branch will receive the given tag(s). For example, say a task is going to collect /intel/perf/foo, /intel/perf/bar, and /intel/perf/baz, all metrics should be tagged with experiment number, additionally one metric /intel/perf/bar should be tagged with OS name. That tags could be described like so:

---
metrics:
  /intel/perf/foo: {}
  /intel/perf/bar: {}
  /intel/perf/baz: {}
tags:
  /intel/perf:
    experiment: "experiment 11"
  /intel/perf/bar:
    os: "linux"

Applying the tags at /intel/perf means that all leaves of /intel/perf (/intel/perf/foo, /intel/perf/bar, and /intel/perf/baz in this case) will receive the tag experiment: experiment 11. Applying the tags at /intel/perf/bar means that only /intel/perf/bar will receive the tag os: linux.

A collect node can also contain any number of process or publish nodes. These nodes describe what to do next.

process

A process node describes which plugin to use to process data coming from either a collection or another process node. The config section describes config data which may be needed for the chosen plugin.

A process node may have any number of process or publish nodes.

publish

A publish node describes which plugin to use to process data coming from either a collection or a process node. The config section describes config data which may be needed for the chosen plugin.

A publish node is a pendant vertex (a leaf). It may contain no collect, process, or publish nodes.

TL;DR

Below is a complete example task.

YAML

---
  version: 1
  schedule:
    type: "simple"
    interval: "1s"
  workflow:
    collect:
      metrics:
        /intel/mock/foo: {}
        /intel/mock/bar: {}
        /intel/mock/*/baz: {}
      config:
        /intel/mock:
          user: "root"
          password: "secret"
      tags:
        /intel/perf:
          experiment: "experiment 11"
        /intel/perf/bar:
          os: "linux"
      process:
        -
          plugin_name: "passthru"
          process: null
          publish:
            -
              plugin_name: "file"
              config:
                file: "/tmp/published"

JSON

{
    "version": 1,
    "schedule": {
        "type": "simple",
        "interval": "1s"
    },
    "workflow": {
        "collect": {
            "metrics": {
                "/intel/mock/foo": {},
                "/intel/mock/bar": {},
                "/intel/mock/*/baz": {}
            },
            "config": {
                "/intel/mock": {
                    "user": "root",
                    "password": "secret"
                }
            },
            "tags": {
                "/intel/mock": {
                    "experiment": "experiment 11"
                },
                "/intel/mock/bar": {
                    "os": "linux"
                }
            },
            "process": [
                {
                    "plugin_name": "passthru",
                    "process": null,
                    "publish": [
                        {
                            "plugin_name": "file",
                            "config": {
                                "file": "/tmp/published"
                            }
                        }
                    ]
                }
            ]
        }
    }
}

footnotes

  1. YAML is only supported via the snaptel CLI. Only JSON is accepted via the REST API.
  2. The wildcard must be supported by the target plugin.