Skip to content

Sherlock is an anomaly detection service built on top of Druid

License

Notifications You must be signed in to change notification settings

mbrtargeting/sherlock

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Sherlock: Anomaly Detector

build Release Artifacts Snapshot Artifacts Coverage Status GPL 3.0

Table of Contents

Introduction to Sherlock

Sherlock is an anomaly detection service built on top of Druid. It leverages EGADS (Extensible Generic Anomaly Detection System) to detect anomalies in time-series data. Users can schedule jobs on an hourly, daily, weekly, or monthly basis, view anomaly reports from Sherlock's interface, or receive them via email.

Components

  1. Timeseries Generation
  2. EGADS Anomaly Detection
  3. Redis database
  4. UI in Spark Java

Detailed Description

Timeseries Generation

Timeseries generation is the first phase of Sherlock's anomaly detection. The user inputs a full Druid JSON query with a metric name and group-by dimensions. Sherlock validates the query, adjusts the time interaval and granularity based on the EGADS config, and makes a call to Druid. Druid responds with an array of time-series, which are parsed into EGADS time-series.

Sample Druid Query:

{
  "metric": "metric(metric1/metric2)", 
  "aggregations": [
    {
      "filter": {
        "fields": [
          {
            "type": "selector", 
            "dimension": "dim1", 
            "value": "value1"
          }
        ], 
        "type": "or"
      }, 
      "aggregator": {
        "fieldName": "metric2", 
        "type": "longSum", 
        "name": "metric2"
      }, 
      "type": "filtered"
    }
  ], 
  "dimension": "groupByDimension", 
  "intervals": "2017-09-10T00:00:01+00:00/2017-10-12T00:00:01+00:00", 
  "dataSource": "source1", 
  "granularity": {
    "timeZone": "UTC", 
    "type": "period", 
    "period": "P1D"
  }, 
  "threshold": 50, 
  "postAggregations": [
    {
      "fields": [
        {
          "fieldName": "metric1", 
          "type": "fieldAccess", 
          "name": "metric1"
        }
      ], 
      "type": "arithmetic", 
      "name": "metric(metric1/metric2)", 
      "fn": "/"
    }
  ], 
  "queryType": "topN"
}

Sample Druid Response:

[ {
  "timestamp" : "2017-10-11T00:00:00.000Z",
  "result" : [ {
    "groupByDimension" : "dim1",
    "metric(metric1/metric2)" : 8,
    "metric1" : 128,
    "metric2" : 16
  }, {
    "groupByDimension" : "dim2",
    "metric(metric1/metric2)" : 4.5,
    "metric1" : 42,
    "metric2" : 9.33
  } ]
}, {
  "timestamp" : "2017-10-12T00:00:00.000Z",
  "result" : [ {
    "groupByDimension" : "dim1",
    "metric(metric1/metric2)" : 9,
    "metric1" : 180,
    "metric2" : 20
  }, {
    "groupByDimension" : "dim2",
    "metric(metric1/metric2)" : 5.5,
    "metric1" : 95,
    "metric2" : 17.27
  } ]
} ]

EGADS Anomaly Detection

Sherlock calls the user-configured EGADS API for each generated time-series, generates anomaly reports from the response, and stores these reports in a database. Users may also elect to receive anomaly reports by email.

Redis Database

Sherlock uses a Redis backend Redis to store job metadata, generated anomaly reports, among other information, and as a persistent job queue. Keys related to Reports have retention policy. Hourly job reports have retention of 14 days and daily/weekly/monthly job reports have 1 year of retention.

Sherlock UI

Sherlock's user interface is built with Spark. The UI enables users to submit instant anomaly analyses, create and launch detection jobs, view anomalies on a heatmap, and on a graph.

Building Sherlock

A Makefile is provided with all build targets.

Building the JAR

make jar

This creates sherlock.jar in the target/ directory.

How to run

Sherlock is run through the commandline with config arguments.

java -Dlog4j.configuration=file:${path_to_log4j}/log4j.properties \
      -jar ${path_to_jar}/sherlock.jar \
      --version $(VERSION) \
      --project-name $(PROJECT_NAME) \
      --port $(PORT) \
      --enable-email \
      --failure-email $(FAILURE_EMAIL) \
      --from-mail $(FROM_MAIL) \
      --reply-to $(REPLY_TO) \
      --smtp-host $(SMTP_HOST) \
      --interval-minutes $(INTERVAL_MINUTES) \
      --interval-hours $(INTERVAL_HOURS) \
      --interval-days $(INTERVAL_DAYS) \
      --interval-weeks $(INTERVAL_WEEKS) \
      --interval-months $(INTERVAL_MONTHS) \
      --egads-config-filename $(EGADS_CONFIG_FILENAME) \
      --redis-host $(REDIS_HOSTNAME) \
      --redis-port $(REDIS_PORT) \
      --execution-delay $(EXECUTION_DELAY) \
      --timeseries-completeness $(TIMESERIES_COMPLETENESS)

CLI args usage

args required default description
--help - false help
--config - null config
--version - v0.0.0 version
--egads-config-filename - provided egads-config-filename
--port - 4080 port
--interval-minutes - 180 interval-minutes
--interval-hours - 672 interval-hours
--interval-days - 28 interval-days
--interval-weeks - 12 interval-weeks
--interval-months - 6 interval-months
--enable-email - false enable-email
--from-mail if email enabled from-mail
--reply-to if email enabled reply-to
--smtp-host if email enabled smtp-host
--smtp-port - 25 smtp-port
--smtp-user - smtp-user
--smtp-password - smtp-password
--failure-email if email enabled failure-email
--execution-delay - 30 execution-delay
--valid-domains - null valid-domains
--redis-host - 127.0.0.1 redis-host
--redis-port - 6379 redis-port
--redis-ssl - false redis-ssl
--redis-timeout - 5000 redis-timeout
--redis-password - - redis-password
--redis-clustered - false redis-clustered
--project-name - - project-name
--external-file-path - - external-file-path
--debug-mode - false debug-mode
--timeseries-completeness - 60 timeseries-completeness
--http-client-timeout - 20000 http-client-timeout
--nodata-on-failure - false nodata-on-failure
--backup-redis-db-path - null backup-redis-db-path
--druid-brokers-list-file - null druid-brokers-list-file
--truststore-path - null truststore-path
--truststore-type - jks truststore-type
--truststore-password - null truststore-password
--keystore-path - null keystore-path
--keystore-type - jks keystore-type
--keystore-password - null keystore-password
--key-dir - null key-dir
--cert-dir - null cert-dir
--https-hostname-verification - true https-hostname-verification
--custom-ssl-context-provider-class - DefaultSslContextProvider custom-ssl-context-provider-class
--custom-secret-provider-class - DefaultSecretProvider custom-secret-provider-class

help

Prints commandline argument help message.

config

Path to a Sherlock configuration file, where the above configuration may be specified. Config arguments in the file override commandline arguments.

version

Version of sherlock.jar to display on the UI

egads-config-filename

Path to a custom EGADS configuration file. If none is specified, the default configuration is used.

port

Port on which to host the Spark application.

interval-minutes

Number of historic data points to use for detection on time-series every minute.

interval-hours

Number of historic data points to use for detection on hourly time-series.

interval-days

Number of historic data points to use for detection on daily time-series.

interval-weeks

Number of historic data points to use for detection on weekly time-series.

interval-months

Number of historic data points to use for detection on monthly time-series.

enable-email

Enable the email service. This enables users to receive email anomaly report notifications.

from-mail

The handle's FROM email displayed to email recipients.

reply-to

The handle's REPLY TO email where replies will be sent.

smtp-host

The email service's SMTP HOST.

smtp-port

The email service's SMTP PORT. The default value is 25.

smtp-user

The email service's SMTP USER.

smtp-password

The email service's SMTP PASSWORD.

failure-email

A dedicated email which may be set to receive job failure notifications.

execution-delay

Sherlock periodically pings Redis to check scheduled jobs. This sets the ping delay in seconds. Jobs are scheduled with a precision of one minute.

valid-domains

A comma-separated list of valid domains to receive emails, e.g. 'yahoo,gmail,hotmail'. If specified, Sherlock will restrict who may receive emails.

redis-host

The Redis backend hostname.

redis-port

The Redis backend port.

redis-ssl

Whether Sherlock should connect to Redis via SSL.

redis-timeout

The Redis connection timeout.

redis-password

The password to use when authenticating to Redis.

redis-clustered

Whether the Redis backend is a cluster.

project-name

Name of the project to display on UI.

external-file-path

Specify the path to external files for Spark framework via this argument.

debug-mode

Debug mode enables debug routes. Ex. '/DatabaseJson' (shows redis data as json dump). Look at com.yahoo.sherlock.App for more details.

timeseries-completeness

This defines minimum fraction of datapoints needed in the timeseries to consider it as a valid timeseries o/w sherlock ignores such timeseries. (default value 60 i.e. 0.6 in fraction)

http-client-timeout

HttpClient timeout can be configured using this(in millis). (default value 20000)

nodata-on-failure

Specify if a job should be set to NODATA rather than ERROR following a druid query failure. (default value false)

backup-redis-db-path

Backup redis DB at given file path as json dump of indices and objects. Backup is done per day at midnight. Default this parameter is null i.e. no buckup. However, BGSAVE command is run at midnight to save redis local dump.

druid-brokers-list-file

Specify the path to an access control list file of permitted druid broker hosts for querying. Format: <host1>:<port>,<host2>:<port>... (default null i.e any host is allowed)

truststore-path

Path to specify truststore location for mTLS connections. (default null)

truststore-type

Param to specify truststore type for mTLS connections. (default jks)

truststore-password

Param to specify truststore password for mTLS connections. (default null)

keystore-path

Path to specify keystore location for mTLS connections. (default null)

keystore-type

Param to specify keystore type for mTLS connections. (default jks)

keystore-password

Param to specify keystore password for mTLS connections. (default null)

key-dir

Param to specify key directory containing multiple keys(for different clusters) for mTLS connections (default null). This is used when Principal Name is given in druid cluster form. It looks for filename containing Principal Name under this dir. If --key-dir and --cert-dir values are same then the filename should also contain the identifier key for private key file and cert for public key file.

cert-dir

Param to specify cert directory containing multiple certs(for different clusters) for mTLS connections (default null)." This is used when Principal Name is given in druid cluster form. It looks for file name containing Principal Name under this dir. If --key-dir and --cert-dir values are same then the filename should also contain the identifier key for private key file and cert for public key file.

https-hostname-verification

Param to enable/disable https hostname verification for mTLS connections. (default true i.e. hostname verification enabled)

custom-ssl-context-provider-class

Param to specify custom ssl context provider class for mTLS connections. (default com.yahoo.sherlock.utils.DefaultSslContextProvider which returns SSLContext with validation)

custom-secret-provider-class

Param to specify custom secret provider class for passwords. (default com.yahoo.sherlock.utils.DefaultSecretProvider which returns secrets specified from CLISettings)

Committers

Jigar Patel, [email protected]

Jeff Niu, [email protected]

Contributors

Josh Walters, [email protected]

Stephan Stiefel, Stephan3555

License

Code licensed under the GPL v3 License. See LICENSE file for terms.

About

Sherlock is an anomaly detection service built on top of Druid

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages

  • Java 56.8%
  • JavaScript 28.4%
  • HTML 12.8%
  • CSS 2.0%