nitpick

This reverts commit 7fa27f400d.

Makefile

@@ -54,11 +54,11 @@ fmtcheck:
 	@echo '[INFO] done.'
 
 test-windows:
-	go test -short ./plugins/inputs/ping/...
-	go test -short ./plugins/inputs/win_perf_counters/...
-	go test -short ./plugins/inputs/win_services/...
-	go test -short ./plugins/inputs/procstat/...
-	go test -short ./plugins/inputs/ntpq/...
+	go test ./plugins/inputs/ping/...
+	go test ./plugins/inputs/win_perf_counters/...
+	go test ./plugins/inputs/win_services/...
+	go test ./plugins/inputs/procstat/...
+	go test ./plugins/inputs/ntpq/...
 
 # vet runs the Go source code static analysis tool `vet` to find
 # any common errors.
@@ -92,4 +92,4 @@ docker-image:
 plugins/parsers/influx/machine.go: plugins/parsers/influx/machine.go.rl
 	ragel -Z -G2 $^ -o $@
 
-.PHONY: deps telegraf install test test-windows lint vet test-all package clean docker-image fmtcheck uint64
+.PHONY: deps telegraf install test test-windows lint vet test-all package clean docker-image fmtcheck uint64

docs/DATA_FORMATS_INPUT.md

@@ -9,6 +9,8 @@ Telegraf is able to parse the following input data formats into metrics:
 1. [Nagios](https://github.com/influxdata/telegraf/blob/master/docs/DATA_FORMATS_INPUT.md#nagios) (exec input only)
 1. [Collectd](https://github.com/influxdata/telegraf/blob/master/docs/DATA_FORMATS_INPUT.md#collectd)
 1. [Dropwizard](https://github.com/influxdata/telegraf/blob/master/docs/DATA_FORMATS_INPUT.md#dropwizard)
+1. [Grok](https://github.com/influxdata/telegraf/blob/master/docs/DATA_FORMATS_INPUT.md#grok)
+
 
 Telegraf metrics, like InfluxDB
 [points](https://docs.influxdata.com/influxdb/v0.10/write_protocols/line/),
@@ -651,5 +653,37 @@ For more information about the dropwizard json format see
   # [inputs.exec.dropwizard_tag_paths]
   #   tag1 = "tags.tag1"
   #   tag2 = "tags.tag2"
+```
 
-```
+#### Grok
+Parse logstash-style "grok" patterns:
+```toml
+  [inputs.reader]
+    ## This is a list of patterns to check the given log file(s) for.
+    ## Note that adding patterns here increases processing time. The most
+    ## efficient configuration is to have one pattern per logparser.
+    ## Other common built-in patterns are:
+    ##   %{COMMON_LOG_FORMAT}   (plain apache & nginx access logs)
+    ##   %{COMBINED_LOG_FORMAT} (access logs + referrer & agent)
+    patterns = ["%{COMBINED_LOG_FORMAT}"]
+
+    ## Name of the outputted measurement name.
+    name_override = "apache_access_log"
+
+    ## Full path(s) to custom pattern files.
+    custom_pattern_files = []
+
+    ## Custom patterns can also be defined here. Put one pattern per line.
+    custom_patterns = '''
+
+    ## Timezone allows you to provide an override for timestamps that
+    ## don't already include an offset
+    ## e.g. 04/06/2016 12:41:45 data one two 5.43µs
+    ##
+    ## Default: "" which renders UTC
+    ## Options are as follows:
+    ##   1. Local             -- interpret based on machine localtime
+    ##   2. "Canada/Eastern"  -- Unix TZ values like those found in https://en.wikipedia.org/wiki/List_of_tz_database_time_zones
+    ##   3. UTC               -- or blank/unspecified, will return timestamp in UTC
+    timezone = "Canada/Eastern"
+```

internal/config/config.go

@@ -211,16 +211,12 @@ var header = `# Telegraf Configuration
 # Environment variables can be used anywhere in this config file, simply prepend
 # them with $. For strings the variable must be within quotes (ie, "$STR_VAR"),
 # for numbers and booleans they should be plain (ie, $INT_VAR, $BOOL_VAR)
-
-
 # Global tags can be specified here in key="value" format.
 [global_tags]
   # dc = "us-east-1" # will tag all metrics with dc=us-east-1
   # rack = "1a"
   ## Environment variables can be used as tags, and throughout the config file
   # user = "$USER"
-
-
 # Configuration for telegraf agent
 [agent]
   ## Default data collection interval for all inputs
@@ -228,24 +224,20 @@ var header = `# Telegraf Configuration
   ## Rounds collection interval to 'interval'
   ## ie, if interval="10s" then always collect on :00, :10, :20, etc.
   round_interval = true
-
   ## Telegraf will send metrics to outputs in batches of at most
   ## metric_batch_size metrics.
   ## This controls the size of writes that Telegraf sends to output plugins.
   metric_batch_size = 1000
-
   ## For failed writes, telegraf will cache metric_buffer_limit metrics for each
   ## output, and will flush this buffer on a successful write. Oldest metrics
   ## are dropped first when this buffer fills.
   ## This buffer only fills when writes fail to output plugin(s).
   metric_buffer_limit = 10000
-
   ## Collection jitter is used to jitter the collection by a random amount.
   ## Each plugin will sleep for a random time within jitter before collecting.
   ## This can be used to avoid many plugins querying things like sysfs at the
   ## same time, which can have a measurable effect on the system.
   collection_jitter = "0s"
-
   ## Default flushing interval for all outputs. You shouldn't set this below
   ## interval. Maximum flush_interval will be flush_interval + flush_jitter
   flush_interval = "10s"
@@ -253,7 +245,6 @@ var header = `# Telegraf Configuration
   ## large write spikes for users running a large number of telegraf instances.
   ## ie, a jitter of 5s and interval 10s means flushes will happen every 10-15s
   flush_jitter = "0s"
-
   ## By default or when set to "0s", precision will be set to the same
   ## timestamp order as the collection interval, with the maximum being 1s.
   ##   ie, when interval = "10s", precision will be "1s"
@@ -262,7 +253,6 @@ var header = `# Telegraf Configuration
   ## service input to set the timestamp at the appropriate precision.
   ## Valid time units are "ns", "us" (or "µs"), "ms", "s".
   precision = ""
-
   ## Logging configuration:
   ## Run telegraf with debug log messages.
   debug = false
@@ -270,41 +260,34 @@ var header = `# Telegraf Configuration
   quiet = false
   ## Specify the log file name. The empty string means to log to stderr.
   logfile = ""
-
   ## Override default hostname, if empty use os.Hostname()
   hostname = ""
   ## If set to true, do no set the "host" tag in the telegraf agent.
   omit_hostname = false
-
-
 ###############################################################################
 #                            OUTPUT PLUGINS                                   #
 ###############################################################################
 `
 
 var processorHeader = `
-
 ###############################################################################
 #                            PROCESSOR PLUGINS                                #
 ###############################################################################
 `
 
 var aggregatorHeader = `
-
 ###############################################################################
 #                            AGGREGATOR PLUGINS                               #
 ###############################################################################
 `
 
 var inputHeader = `
-
 ###############################################################################
 #                            INPUT PLUGINS                                    #
 ###############################################################################
 `
 
 var serviceInputHeader = `
-
 ###############################################################################
 #                            SERVICE INPUT PLUGINS                            #
 ###############################################################################
@@ -1338,6 +1321,59 @@ func buildParser(name string, tbl *ast.Table) (parsers.Parser, error) {
 		}
 	}
 
+	//for grok data_format
+	if node, ok := tbl.Fields["named_patterns"]; ok {
+		if kv, ok := node.(*ast.KeyValue); ok {
+			if ary, ok := kv.Value.(*ast.Array); ok {
+				for _, elem := range ary.Value {
+					if str, ok := elem.(*ast.String); ok {
+						c.NamedPatterns = append(c.NamedPatterns, str.Value)
+					}
+				}
+			}
+		}
+	}
+
+	if node, ok := tbl.Fields["patterns"]; ok {
+		if kv, ok := node.(*ast.KeyValue); ok {
+			if ary, ok := kv.Value.(*ast.Array); ok {
+				for _, elem := range ary.Value {
+					if str, ok := elem.(*ast.String); ok {
+						c.Patterns = append(c.Patterns, str.Value)
+					}
+				}
+			}
+		}
+	}
+
+	if node, ok := tbl.Fields["custom_patterns"]; ok {
+		if kv, ok := node.(*ast.KeyValue); ok {
+			if str, ok := kv.Value.(*ast.String); ok {
+				c.CustomPatterns = str.Value
+			}
+		}
+	}
+
+	if node, ok := tbl.Fields["custom_pattern_files"]; ok {
+		if kv, ok := node.(*ast.KeyValue); ok {
+			if ary, ok := kv.Value.(*ast.Array); ok {
+				for _, elem := range ary.Value {
+					if str, ok := elem.(*ast.String); ok {
+						c.CustomPatternFiles = append(c.CustomPatternFiles, str.Value)
+					}
+				}
+			}
+		}
+	}
+
+	if node, ok := tbl.Fields["timezone"]; ok {
+		if kv, ok := node.(*ast.KeyValue); ok {
+			if str, ok := kv.Value.(*ast.String); ok {
+				c.TimeZone = str.Value
+			}
+		}
+	}
+
 	c.MetricName = name
 
 	delete(tbl.Fields, "data_format")
@@ -1353,6 +1389,11 @@ func buildParser(name string, tbl *ast.Table) (parsers.Parser, error) {
 	delete(tbl.Fields, "dropwizard_time_format")
 	delete(tbl.Fields, "dropwizard_tags_path")
 	delete(tbl.Fields, "dropwizard_tag_paths")
+	delete(tbl.Fields, "named_patterns")
+	delete(tbl.Fields, "patterns")
+	delete(tbl.Fields, "custom_patterns")
+	delete(tbl.Fields, "custom_pattern_files")
+	delete(tbl.Fields, "timezone")
 
 	return parsers.NewParser(c)
 }

plugins/inputs/all/all.go

@@ -85,6 +85,7 @@ import (
 	_ "github.com/influxdata/telegraf/plugins/inputs/puppetagent"
 	_ "github.com/influxdata/telegraf/plugins/inputs/rabbitmq"
 	_ "github.com/influxdata/telegraf/plugins/inputs/raindrops"
+	_ "github.com/influxdata/telegraf/plugins/inputs/reader"
 	_ "github.com/influxdata/telegraf/plugins/inputs/redis"
 	_ "github.com/influxdata/telegraf/plugins/inputs/rethinkdb"
 	_ "github.com/influxdata/telegraf/plugins/inputs/riak"

plugins/inputs/reader/README.md

@@ -0,0 +1,23 @@
+# Reader Input Plugin
+
+The `reader` plugin reads and parses files every interval.  Reader will always begin at the top of each file.
+
+Reader supports all data_format formats
+
+### Configuration
+
+```toml
+## Files to parse each interval.
+## These accept standard unix glob matching rules, but with the addition of
+## ** as a "super asterisk". ie:
+##   /var/log/**.log     -> recursively find all .log files in /var/log
+##   /var/log/*/*.log    -> find all .log files with a parent dir in /var/log
+##   /var/log/apache.log -> only tail the apache log file
+files = ["/var/log/apache/access.log"]
+
+## The dataformat to be read from files
+## Each data format has its own unique set of configuration options, read
+## more about them here:
+## https://github.com/influxdata/telegraf/blob/master/docs/DATA_FORMATS_INPUT.md
+data_format = ""
+```

plugins/inputs/reader/dev/docker-compose.yml

@@ -0,0 +1,14 @@
+version: '3'
+
+services:
+  telegraf:
+      image: glinton/scratch
+      volumes:
+        - ./telegraf.conf:/telegraf.conf
+        - ../../../../telegraf:/telegraf
+        - ./json_a.log:/var/log/test.log
+      entrypoint:
+        - /telegraf
+        - --config
+        - /telegraf.conf
+        

plugins/inputs/reader/dev/json_a.log

@@ -0,0 +1,15 @@
+{
+		"parent": {
+			"child": 3.0,
+			"ignored_child": "hi"
+		},
+		"ignored_null": null,
+		"integer": 4,
+		"list": [3, 4],
+		"ignored_parent": {
+			"another_ignored_null": null,
+			"ignored_string": "hello, world!"
+		},
+		"another_list": [4]
+	}
+	

plugins/inputs/reader/dev/telegraf.conf

@@ -0,0 +1,8 @@
+[[inputs.reader]]
+  files = ["/var/log/test.log"]
+  data_format = "json"
+  name_override = "json_reader"
+
+[[outputs.file]]
+  files = ["stdout"]
+  

plugins/inputs/reader/reader.go

@@ -0,0 +1,102 @@
+package reader
+
+import (
+	"io/ioutil"
+	"log"
+
+	"github.com/influxdata/telegraf"
+	"github.com/influxdata/telegraf/internal/globpath"
+	"github.com/influxdata/telegraf/plugins/inputs"
+	"github.com/influxdata/telegraf/plugins/parsers"
+)
+
+type Reader struct {
+	Filepaths     []string `toml:"files"`
+	FromBeginning bool
+	parser        parsers.Parser
+
+	Filenames []string
+}
+
+const sampleConfig = `## Files to parse each interval.
+## These accept standard unix glob matching rules, but with the addition of
+## ** as a "super asterisk". ie:
+##   /var/log/**.log     -> recursively find all .log files in /var/log
+##   /var/log/*/*.log    -> find all .log files with a parent dir in /var/log
+##   /var/log/apache.log -> only tail the apache log file
+files = ["/var/log/apache/access.log"]
+
+## The dataformat to be read from files
+## Each data format has its own unique set of configuration options, read
+## more about them here:
+## https://github.com/influxdata/telegraf/blob/master/docs/DATA_FORMATS_INPUT.md
+data_format = ""
+`
+
+// SampleConfig returns the default configuration of the Input
+func (r *Reader) SampleConfig() string {
+	return sampleConfig
+}
+
+func (r *Reader) Description() string {
+	return "reload and gather from file[s] on telegraf's interval"
+}
+
+func (r *Reader) Gather(acc telegraf.Accumulator) error {
+	r.refreshFilePaths()
+	for _, k := range r.Filenames {
+		metrics, err := r.readMetric(k)
+		if err != nil {
+			return err
+		}
+
+		for i, m := range metrics {
+
+			//error if m is nil
+			if m == nil {
+				log.Printf("E! Metric could not be parsed from: %v, on line %v", k, i)
+				continue
+			}
+			acc.AddFields(m.Name(), m.Fields(), m.Tags())
+		}
+	}
+	return nil
+}
+
+func (r *Reader) SetParser(p parsers.Parser) {
+	r.parser = p
+}
+
+func (r *Reader) refreshFilePaths() {
+	var allFiles []string
+	for _, filepath := range r.Filepaths {
+		g, err := globpath.Compile(filepath)
+		if err != nil {
+			log.Printf("E! Error Glob %s failed to compile, %s", filepath, err)
+			continue
+		}
+		files := g.Match()
+
+		for k := range files {
+			allFiles = append(allFiles, k)
+		}
+	}
+
+	r.Filenames = allFiles
+}
+
+//requires that Parser has been compiled
+func (r *Reader) readMetric(filename string) ([]telegraf.Metric, error) {
+	fileContents, err := ioutil.ReadFile(filename)
+	if err != nil {
+		log.Printf("E! File could not be opened: %v", filename)
+	}
+	return r.parser.Parse(fileContents)
+
+}
+
+func init() {
+	inputs.Add("reader", func() telegraf.Input {
+		return &Reader{}
+	})
+}

plugins/inputs/reader/reader_test.go

@@ -0,0 +1,64 @@
+package reader
+
+import (
+	"runtime"
+	"strings"
+	"testing"
+
+	"github.com/influxdata/telegraf/plugins/parsers"
+	"github.com/influxdata/telegraf/testutil"
+	"github.com/stretchr/testify/assert"
+)
+
+func TestRefreshFilePaths(t *testing.T) {
+	testDir := getPluginDir()
+	r := Reader{
+		Filepaths: []string{testDir + "/logparser/grok/testdata/**.log"},
+	}
+
+	r.refreshFilePaths()
+	assert.Equal(t, len(r.Filenames), 2)
+}
+func TestJSONParserCompile(t *testing.T) {
+	testDir := getPluginDir()
+	var acc testutil.Accumulator
+	r := Reader{
+		Filepaths: []string{testDir + "/reader/testfiles/json_a.log"},
+	}
+	parserConfig := parsers.Config{
+		DataFormat: "json",
+		TagKeys:    []string{"parent_ignored_child"},
+	}
+	nParser, err := parsers.NewParser(&parserConfig)
+	r.parser = nParser
+	assert.NoError(t, err)
+
+	r.Gather(&acc)
+	assert.Equal(t, map[string]string{"parent_ignored_child": "hi"}, acc.Metrics[0].Tags)
+	assert.Equal(t, 5, len(acc.Metrics[0].Fields))
+}
+
+func TestGrokParser(t *testing.T) {
+	testDir := getPluginDir()
+	var acc testutil.Accumulator
+	r := Reader{
+		Filepaths: []string{testDir + "/reader/testfiles/grok_a.log"},
+	}
+
+	parserConfig := parsers.Config{
+		DataFormat: "grok",
+		Patterns:   []string{"%{COMMON_LOG_FORMAT}"},
+	}
+
+	nParser, err := parsers.NewParser(&parserConfig)
+	r.parser = nParser
+	assert.NoError(t, err)
+
+	err = r.Gather(&acc)
+	assert.Equal(t, 2, len(acc.Metrics))
+}
+
+func getPluginDir() string {
+	_, filename, _, _ := runtime.Caller(1)
+	return strings.Replace(filename, "/reader/reader_test.go", "", 1)
+}

plugins/inputs/reader/testfiles/grok_a.log

@@ -0,0 +1,2 @@
+127.0.0.1 user-identifier frank [10/Oct/2000:13:55:36 -0700] "GET /apache_pb.gif HTTP/1.0" 200 2326
+128.0.0.1 user-identifier tony [10/Oct/2000:13:55:36 -0800] "GET /apache_pb.gif HTTP/1.0" 300 45

plugins/inputs/reader/testfiles/json_a.log

@@ -0,0 +1,14 @@
+{
+		"parent": {
+			"child": 3.0,
+			"ignored_child": "hi"
+		},
+		"ignored_null": null,
+		"integer": 4,
+		"list": [3, 4],
+		"ignored_parent": {
+			"another_ignored_null": null,
+			"ignored_string": "hello, world!"
+		},
+		"another_list": [4]
+	}

plugins/parsers/grok/influx-patterns

@@ -0,0 +1,73 @@
+# Captures are a slightly modified version of logstash "grok" patterns, with
+#  the format %{<capture syntax>[:<semantic name>][:<modifier>]}
+# By default all named captures are converted into string fields.
+# Modifiers can be used to convert captures to other types or tags.
+# Timestamp modifiers can be used to convert captures to the timestamp of the
+#  parsed metric.
+
+# View logstash grok pattern docs here:
+#   https://www.elastic.co/guide/en/logstash/current/plugins-filters-grok.html
+# All default logstash patterns are supported, these can be viewed here:
+#   https://github.com/logstash-plugins/logstash-patterns-core/blob/master/patterns/grok-patterns
+
+# Available modifiers:
+#   string   (default if nothing is specified)
+#   int
+#   float
+#   duration (ie, 5.23ms gets converted to int nanoseconds)
+#   tag      (converts the field into a tag)
+#   drop     (drops the field completely)
+# Timestamp modifiers:
+#   ts-ansic         ("Mon Jan _2 15:04:05 2006")
+#   ts-unix          ("Mon Jan _2 15:04:05 MST 2006")
+#   ts-ruby          ("Mon Jan 02 15:04:05 -0700 2006")
+#   ts-rfc822        ("02 Jan 06 15:04 MST")
+#   ts-rfc822z       ("02 Jan 06 15:04 -0700")
+#   ts-rfc850        ("Monday, 02-Jan-06 15:04:05 MST")
+#   ts-rfc1123       ("Mon, 02 Jan 2006 15:04:05 MST")
+#   ts-rfc1123z      ("Mon, 02 Jan 2006 15:04:05 -0700")
+#   ts-rfc3339       ("2006-01-02T15:04:05Z07:00")
+#   ts-rfc3339nano   ("2006-01-02T15:04:05.999999999Z07:00")
+#   ts-httpd         ("02/Jan/2006:15:04:05 -0700")
+#   ts-epoch         (seconds since unix epoch)
+#   ts-epochnano     (nanoseconds since unix epoch)
+#   ts-"CUSTOM"
+# CUSTOM time layouts must be within quotes and be the representation of the
+# "reference time", which is Mon Jan 2 15:04:05 -0700 MST 2006
+# See https://golang.org/pkg/time/#Parse for more details.
+
+# Example log file pattern, example log looks like this:
+#   [04/Jun/2016:12:41:45 +0100] 1.25 200 192.168.1.1 5.432µs
+# Breakdown of the DURATION pattern below:
+#   NUMBER  is a builtin logstash grok pattern matching float & int numbers.
+#   [nuµm]? is a regex specifying 0 or 1 of the characters within brackets.
+#   s       is also regex, this pattern must end in "s".
+# so DURATION will match something like '5.324ms' or '6.1µs' or '10s'
+DURATION %{NUMBER}[nuµm]?s
+RESPONSE_CODE %{NUMBER:response_code:tag}
+RESPONSE_TIME %{DURATION:response_time_ns:duration}
+EXAMPLE_LOG \[%{HTTPDATE:ts:ts-httpd}\] %{NUMBER:myfloat:float} %{RESPONSE_CODE} %{IPORHOST:clientip} %{RESPONSE_TIME}
+
+# Wider-ranging username matching vs. logstash built-in %{USER}
+NGUSERNAME [a-zA-Z0-9\.\@\-\+_%]+
+NGUSER %{NGUSERNAME}
+# Wider-ranging client IP matching
+CLIENT (?:%{IPORHOST}|%{HOSTPORT}|::1)
+
+##
+## COMMON LOG PATTERNS
+##
+
+# apache & nginx logs, this is also known as the "common log format"
+#   see https://en.wikipedia.org/wiki/Common_Log_Format
+COMMON_LOG_FORMAT %{CLIENT:client_ip} %{NOTSPACE:ident} %{NOTSPACE:auth} \[%{HTTPDATE:ts:ts-httpd}\] "(?:%{WORD:verb:tag} %{NOTSPACE:request}(?: HTTP/%{NUMBER:http_version:float})?|%{DATA})" %{NUMBER:resp_code:tag} (?:%{NUMBER:resp_bytes:int}|-)
+
+# Combined log format is the same as the common log format but with the addition
+# of two quoted strings at the end for "referrer" and "agent"
+#   See Examples at http://httpd.apache.org/docs/current/mod/mod_log_config.html
+COMBINED_LOG_FORMAT %{COMMON_LOG_FORMAT} %{QS:referrer} %{QS:agent}
+
+# HTTPD log formats
+HTTPD20_ERRORLOG \[%{HTTPDERROR_DATE:timestamp}\] \[%{LOGLEVEL:loglevel:tag}\] (?:\[client %{IPORHOST:clientip}\] ){0,1}%{GREEDYDATA:errormsg}
+HTTPD24_ERRORLOG \[%{HTTPDERROR_DATE:timestamp}\] \[%{WORD:module}:%{LOGLEVEL:loglevel:tag}\] \[pid %{POSINT:pid:int}:tid %{NUMBER:tid:int}\]( \(%{POSINT:proxy_errorcode:int}\)%{DATA:proxy_errormessage}:)?( \[client %{IPORHOST:client}:%{POSINT:clientport}\])? %{DATA:errorcode}: %{GREEDYDATA:message}
+HTTPD_ERRORLOG %{HTTPD20_ERRORLOG}|%{HTTPD24_ERRORLOG}

plugins/parsers/grok/influx_patterns.go

@@ -0,0 +1,78 @@
+package grok
+
+// DEFAULT_PATTERNS SHOULD BE KEPT IN-SYNC WITH patterns/influx-patterns
+const DEFAULT_PATTERNS = `
+# Captures are a slightly modified version of logstash "grok" patterns, with
+#  the format %{<capture syntax>[:<semantic name>][:<modifier>]}
+# By default all named captures are converted into string fields.
+# Modifiers can be used to convert captures to other types or tags.
+# Timestamp modifiers can be used to convert captures to the timestamp of the
+#  parsed metric.
+
+# View logstash grok pattern docs here:
+#   https://www.elastic.co/guide/en/logstash/current/plugins-filters-grok.html
+# All default logstash patterns are supported, these can be viewed here:
+#   https://github.com/logstash-plugins/logstash-patterns-core/blob/master/patterns/grok-patterns
+
+# Available modifiers:
+#   string   (default if nothing is specified)
+#   int
+#   float
+#   duration (ie, 5.23ms gets converted to int nanoseconds)
+#   tag      (converts the field into a tag)
+#   drop     (drops the field completely)
+# Timestamp modifiers:
+#   ts-ansic         ("Mon Jan _2 15:04:05 2006")
+#   ts-unix          ("Mon Jan _2 15:04:05 MST 2006")
+#   ts-ruby          ("Mon Jan 02 15:04:05 -0700 2006")
+#   ts-rfc822        ("02 Jan 06 15:04 MST")
+#   ts-rfc822z       ("02 Jan 06 15:04 -0700")
+#   ts-rfc850        ("Monday, 02-Jan-06 15:04:05 MST")
+#   ts-rfc1123       ("Mon, 02 Jan 2006 15:04:05 MST")
+#   ts-rfc1123z      ("Mon, 02 Jan 2006 15:04:05 -0700")
+#   ts-rfc3339       ("2006-01-02T15:04:05Z07:00")
+#   ts-rfc3339nano   ("2006-01-02T15:04:05.999999999Z07:00")
+#   ts-httpd         ("02/Jan/2006:15:04:05 -0700")
+#   ts-epoch         (seconds since unix epoch)
+#   ts-epochnano     (nanoseconds since unix epoch)
+#   ts-"CUSTOM"
+# CUSTOM time layouts must be within quotes and be the representation of the
+# "reference time", which is Mon Jan 2 15:04:05 -0700 MST 2006
+# See https://golang.org/pkg/time/#Parse for more details.
+
+# Example log file pattern, example log looks like this:
+#   [04/Jun/2016:12:41:45 +0100] 1.25 200 192.168.1.1 5.432µs
+# Breakdown of the DURATION pattern below:
+#   NUMBER  is a builtin logstash grok pattern matching float & int numbers.
+#   [nuµm]? is a regex specifying 0 or 1 of the characters within brackets.
+#   s       is also regex, this pattern must end in "s".
+# so DURATION will match something like '5.324ms' or '6.1µs' or '10s'
+DURATION %{NUMBER}[nuµm]?s
+RESPONSE_CODE %{NUMBER:response_code:tag}
+RESPONSE_TIME %{DURATION:response_time_ns:duration}
+EXAMPLE_LOG \[%{HTTPDATE:ts:ts-httpd}\] %{NUMBER:myfloat:float} %{RESPONSE_CODE} %{IPORHOST:clientip} %{RESPONSE_TIME}
+
+# Wider-ranging username matching vs. logstash built-in %{USER}
+NGUSERNAME [a-zA-Z0-9\.\@\-\+_%]+
+NGUSER %{NGUSERNAME}
+# Wider-ranging client IP matching
+CLIENT (?:%{IPV6}|%{IPV4}|%{HOSTNAME}|%{HOSTPORT})
+
+##
+## COMMON LOG PATTERNS
+##
+
+# apache & nginx logs, this is also known as the "common log format"
+#   see https://en.wikipedia.org/wiki/Common_Log_Format
+COMMON_LOG_FORMAT %{CLIENT:client_ip} %{NOTSPACE:ident} %{NOTSPACE:auth} \[%{HTTPDATE:ts:ts-httpd}\] "(?:%{WORD:verb:tag} %{NOTSPACE:request}(?: HTTP/%{NUMBER:http_version:float})?|%{DATA})" %{NUMBER:resp_code:tag} (?:%{NUMBER:resp_bytes:int}|-)
+
+# Combined log format is the same as the common log format but with the addition
+# of two quoted strings at the end for "referrer" and "agent"
+#   See Examples at http://httpd.apache.org/docs/current/mod/mod_log_config.html
+COMBINED_LOG_FORMAT %{COMMON_LOG_FORMAT} %{QS:referrer} %{QS:agent}
+
+# HTTPD log formats
+HTTPD20_ERRORLOG \[%{HTTPDERROR_DATE:timestamp}\] \[%{LOGLEVEL:loglevel:tag}\] (?:\[client %{IPORHOST:clientip}\] ){0,1}%{GREEDYDATA:errormsg}
+HTTPD24_ERRORLOG \[%{HTTPDERROR_DATE:timestamp}\] \[%{WORD:module}:%{LOGLEVEL:loglevel:tag}\] \[pid %{POSINT:pid:int}:tid %{NUMBER:tid:int}\]( \(%{POSINT:proxy_errorcode:int}\)%{DATA:proxy_errormessage}:)?( \[client %{IPORHOST:client}:%{POSINT:clientport}\])? %{DATA:errorcode}: %{GREEDYDATA:message}
+HTTPD_ERRORLOG %{HTTPD20_ERRORLOG}|%{HTTPD24_ERRORLOG}
+`

plugins/parsers/grok/parser.go

@@ -0,0 +1,527 @@
+package grok
+
+import (
+	"bufio"
+	"fmt"
+	"log"
+	"os"
+	"regexp"
+	"strconv"
+	"strings"
+	"time"
+
+	"github.com/vjeantet/grok"
+
+	"github.com/influxdata/telegraf"
+	"github.com/influxdata/telegraf/metric"
+)
+
+var timeLayouts = map[string]string{
+	"ts-ansic":       "Mon Jan _2 15:04:05 2006",
+	"ts-unix":        "Mon Jan _2 15:04:05 MST 2006",
+	"ts-ruby":        "Mon Jan 02 15:04:05 -0700 2006",
+	"ts-rfc822":      "02 Jan 06 15:04 MST",
+	"ts-rfc822z":     "02 Jan 06 15:04 -0700", // RFC822 with numeric zone
+	"ts-rfc850":      "Monday, 02-Jan-06 15:04:05 MST",
+	"ts-rfc1123":     "Mon, 02 Jan 2006 15:04:05 MST",
+	"ts-rfc1123z":    "Mon, 02 Jan 2006 15:04:05 -0700", // RFC1123 with numeric zone
+	"ts-rfc3339":     "2006-01-02T15:04:05Z07:00",
+	"ts-rfc3339nano": "2006-01-02T15:04:05.999999999Z07:00",
+	"ts-httpd":       "02/Jan/2006:15:04:05 -0700",
+	// These three are not exactly "layouts", but they are special cases that
+	// will get handled in the ParseLine function.
+	"ts-epoch":     "EPOCH",
+	"ts-epochnano": "EPOCH_NANO",
+	"ts-syslog":    "SYSLOG_TIMESTAMP",
+	"ts":           "GENERIC_TIMESTAMP", // try parsing all known timestamp layouts.
+}
+
+const (
+	INT               = "int"
+	TAG               = "tag"
+	FLOAT             = "float"
+	STRING            = "string"
+	DURATION          = "duration"
+	DROP              = "drop"
+	EPOCH             = "EPOCH"
+	EPOCH_NANO        = "EPOCH_NANO"
+	SYSLOG_TIMESTAMP  = "SYSLOG_TIMESTAMP"
+	GENERIC_TIMESTAMP = "GENERIC_TIMESTAMP"
+)
+
+var (
+	// matches named captures that contain a modifier.
+	//   ie,
+	//     %{NUMBER:bytes:int}
+	//     %{IPORHOST:clientip:tag}
+	//     %{HTTPDATE:ts1:ts-http}
+	//     %{HTTPDATE:ts2:ts-"02 Jan 06 15:04"}
+	modifierRe = regexp.MustCompile(`%{\w+:(\w+):(ts-".+"|t?s?-?\w+)}`)
+	// matches a plain pattern name. ie, %{NUMBER}
+	patternOnlyRe = regexp.MustCompile(`%{(\w+)}`)
+)
+
+// Parser is the primary struct to handle and grok-patterns defined in the config toml
+type Parser struct {
+	Patterns []string
+	// namedPatterns is a list of internally-assigned names to the patterns
+	// specified by the user in Patterns.
+	// They will look like:
+	//   GROK_INTERNAL_PATTERN_0, GROK_INTERNAL_PATTERN_1, etc.
+	NamedPatterns      []string
+	CustomPatterns     string
+	CustomPatternFiles []string
+	Measurement        string
+
+	// Timezone is an optional component to help render log dates to
+	// your chosen zone.
+	// Default: "" which renders UTC
+	// Options are as follows:
+	// 1. Local             -- interpret based on machine localtime
+	// 2. "America/Chicago" -- Unix TZ values like those found in https://en.wikipedia.org/wiki/List_of_tz_database_time_zones
+	// 3. UTC               -- or blank/unspecified, will return timestamp in UTC
+	Timezone string
+	loc      *time.Location
+
+	// typeMap is a map of patterns -> capture name -> modifier,
+	//   ie, {
+	//          "%{TESTLOG}":
+	//             {
+	//                "bytes": "int",
+	//                "clientip": "tag"
+	//             }
+	//       }
+	typeMap map[string]map[string]string
+	// tsMap is a map of patterns -> capture name -> timestamp layout.
+	//   ie, {
+	//          "%{TESTLOG}":
+	//             {
+	//                "httptime": "02/Jan/2006:15:04:05 -0700"
+	//             }
+	//       }
+	tsMap map[string]map[string]string
+	// patterns is a map of all of the parsed patterns from CustomPatterns
+	// and CustomPatternFiles.
+	//   ie, {
+	//          "DURATION":      "%{NUMBER}[nuµm]?s"
+	//          "RESPONSE_CODE": "%{NUMBER:rc:tag}"
+	//       }
+	patterns map[string]string
+	// foundTsLayouts is a slice of timestamp patterns that have been found
+	// in the log lines. This slice gets updated if the user uses the generic
+	// 'ts' modifier for timestamps. This slice is checked first for matches,
+	// so that previously-matched layouts get priority over all other timestamp
+	// layouts.
+	foundTsLayouts []string
+
+	timeFunc func() time.Time
+	g        *grok.Grok
+	tsModder *tsModder
+}
+
+// Compile is a bound method to Parser which will process the options for our parser
+func (p *Parser) Compile() error {
+	p.typeMap = make(map[string]map[string]string)
+	p.tsMap = make(map[string]map[string]string)
+	p.patterns = make(map[string]string)
+	p.tsModder = &tsModder{}
+	var err error
+	p.g, err = grok.NewWithConfig(&grok.Config{NamedCapturesOnly: true})
+	if err != nil {
+		return err
+	}
+
+	// Give Patterns fake names so that they can be treated as named
+	// "custom patterns"
+	p.NamedPatterns = make([]string, 0, len(p.Patterns))
+	for i, pattern := range p.Patterns {
+		pattern = strings.TrimSpace(pattern)
+		if pattern == "" {
+			continue
+		}
+		name := fmt.Sprintf("GROK_INTERNAL_PATTERN_%d", i)
+		p.CustomPatterns += "\n" + name + " " + pattern + "\n"
+		p.NamedPatterns = append(p.NamedPatterns, "%{"+name+"}")
+	}
+
+	if len(p.NamedPatterns) == 0 {
+		return fmt.Errorf("pattern required")
+	}
+
+	// Combine user-supplied CustomPatterns with DEFAULT_PATTERNS and parse
+	// them together as the same type of pattern.
+	p.CustomPatterns = DEFAULT_PATTERNS + p.CustomPatterns
+	if len(p.CustomPatterns) != 0 {
+		scanner := bufio.NewScanner(strings.NewReader(p.CustomPatterns))
+		p.addCustomPatterns(scanner)
+	}
+
+	// Parse any custom pattern files supplied.
+	for _, filename := range p.CustomPatternFiles {
+		file, fileErr := os.Open(filename)
+		if fileErr != nil {
+			return fileErr
+		}
+
+		scanner := bufio.NewScanner(bufio.NewReader(file))
+		p.addCustomPatterns(scanner)
+	}
+
+	if p.Measurement == "" {
+		p.Measurement = "logparser_grok"
+	}
+
+	p.loc, err = time.LoadLocation(p.Timezone)
+	if err != nil {
+		log.Printf("W! improper timezone supplied (%s), setting loc to UTC", p.Timezone)
+		p.loc, _ = time.LoadLocation("UTC")
+	}
+
+	if p.timeFunc == nil {
+		p.timeFunc = time.Now
+	}
+
+	return p.compileCustomPatterns()
+}
+
+// ParseLine is the primary function to process individual lines, returning the metrics
+func (p *Parser) ParseLine(line string) (telegraf.Metric, error) {
+	var err error
+	// values are the parsed fields from the log line
+	var values map[string]string
+	// the matching pattern string
+	var patternName string
+	for _, pattern := range p.NamedPatterns {
+		if values, err = p.g.Parse(pattern, line); err != nil {
+			return nil, err
+		}
+		if len(values) != 0 {
+			patternName = pattern
+			break
+		}
+	}
+
+	if len(values) == 0 {
+		log.Printf("D! Grok no match found for: %q", line)
+		return nil, nil
+	}
+
+	fields := make(map[string]interface{})
+	tags := make(map[string]string)
+	timestamp := time.Now()
+	for k, v := range values {
+		if k == "" || v == "" {
+			continue
+		}
+
+		// t is the modifier of the field
+		var t string
+		// check if pattern has some modifiers
+		if types, ok := p.typeMap[patternName]; ok {
+			t = types[k]
+		}
+		// if we didn't find a modifier, check if we have a timestamp layout
+		if t == "" {
+			if ts, ok := p.tsMap[patternName]; ok {
+				// check if the modifier is a timestamp layout
+				if layout, ok := ts[k]; ok {
+					t = layout
+				}
+			}
+		}
+		// if we didn't find a type OR timestamp modifier, assume string
+		if t == "" {
+			t = STRING
+		}
+
+		switch t {
+		case INT:
+			iv, err := strconv.ParseInt(v, 10, 64)
+			if err != nil {
+				log.Printf("E! Error parsing %s to int: %s", v, err)
+			} else {
+				fields[k] = iv
+			}
+		case FLOAT:
+			fv, err := strconv.ParseFloat(v, 64)
+			if err != nil {
+				log.Printf("E! Error parsing %s to float: %s", v, err)
+			} else {
+				fields[k] = fv
+			}
+		case DURATION:
+			d, err := time.ParseDuration(v)
+			if err != nil {
+				log.Printf("E! Error parsing %s to duration: %s", v, err)
+			} else {
+				fields[k] = int64(d)
+			}
+		case TAG:
+			tags[k] = v
+		case STRING:
+			fields[k] = strings.Trim(v, `"`)
+		case EPOCH:
+			parts := strings.SplitN(v, ".", 2)
+			if len(parts) == 0 {
+				log.Printf("E! Error parsing %s to timestamp: %s", v, err)
+				break
+			}
+
+			sec, err := strconv.ParseInt(parts[0], 10, 64)
+			if err != nil {
+				log.Printf("E! Error parsing %s to timestamp: %s", v, err)
+				break
+			}
+			ts := time.Unix(sec, 0)
+
+			if len(parts) == 2 {
+				padded := fmt.Sprintf("%-9s", parts[1])
+				nsString := strings.Replace(padded[:9], " ", "0", -1)
+				nanosec, err := strconv.ParseInt(nsString, 10, 64)
+				if err != nil {
+					log.Printf("E! Error parsing %s to timestamp: %s", v, err)
+					break
+				}
+				ts = ts.Add(time.Duration(nanosec) * time.Nanosecond)
+			}
+			timestamp = ts
+		case EPOCH_NANO:
+			iv, err := strconv.ParseInt(v, 10, 64)
+			if err != nil {
+				log.Printf("E! Error parsing %s to int: %s", v, err)
+			} else {
+				timestamp = time.Unix(0, iv)
+			}
+		case SYSLOG_TIMESTAMP:
+			ts, err := time.ParseInLocation("Jan 02 15:04:05", v, p.loc)
+			if err == nil {
+				if ts.Year() == 0 {
+					ts = ts.AddDate(timestamp.Year(), 0, 0)
+				}
+				timestamp = ts
+			} else {
+				log.Printf("E! Error parsing %s to time layout [%s]: %s", v, t, err)
+			}
+		case GENERIC_TIMESTAMP:
+			var foundTs bool
+			// first try timestamp layouts that we've already found
+			for _, layout := range p.foundTsLayouts {
+				ts, err := time.ParseInLocation(layout, v, p.loc)
+				if err == nil {
+					timestamp = ts
+					foundTs = true
+					break
+				}
+			}
+			// if we haven't found a timestamp layout yet, try all timestamp
+			// layouts.
+			if !foundTs {
+				for _, layout := range timeLayouts {
+					ts, err := time.ParseInLocation(layout, v, p.loc)
+					if err == nil {
+						timestamp = ts
+						foundTs = true
+						p.foundTsLayouts = append(p.foundTsLayouts, layout)
+						break
+					}
+				}
+			}
+			// if we still haven't found a timestamp layout, log it and we will
+			// just use time.Now()
+			if !foundTs {
+				log.Printf("E! Error parsing timestamp [%s], could not find any "+
+					"suitable time layouts.", v)
+			}
+		case DROP:
+		// goodbye!
+		default:
+			ts, err := time.ParseInLocation(t, v, p.loc)
+			if err == nil {
+				timestamp = ts
+			} else {
+				log.Printf("E! Error parsing %s to time layout [%s]: %s", v, t, err)
+			}
+		}
+	}
+
+	if len(fields) == 0 {
+		return nil, fmt.Errorf("logparser_grok: must have one or more fields")
+	}
+
+	return metric.New(p.Measurement, tags, fields, p.tsModder.tsMod(timestamp))
+}
+
+func (p *Parser) Parse(buf []byte) ([]telegraf.Metric, error) {
+	lines := strings.Split(string(buf), "\n")
+	var metrics []telegraf.Metric
+
+	for _, line := range lines {
+		m, err := p.ParseLine(line)
+		if err != nil {
+			return nil, err
+		}
+		metrics = append(metrics, m)
+	}
+
+	return metrics, nil
+}
+
+func (p *Parser) SetDefaultTags(tags map[string]string) {
+	//needs implementation
+}
+
+func (p *Parser) addCustomPatterns(scanner *bufio.Scanner) {
+	for scanner.Scan() {
+		line := strings.TrimSpace(scanner.Text())
+		if len(line) > 0 && line[0] != '#' {
+			names := strings.SplitN(line, " ", 2)
+			p.patterns[names[0]] = names[1]
+		}
+	}
+}
+
+func (p *Parser) compileCustomPatterns() error {
+	var err error
+	// check if the pattern contains a subpattern that is already defined
+	// replace it with the subpattern for modifier inheritance.
+	for i := 0; i < 2; i++ {
+		for name, pattern := range p.patterns {
+			subNames := patternOnlyRe.FindAllStringSubmatch(pattern, -1)
+			for _, subName := range subNames {
+				if subPattern, ok := p.patterns[subName[1]]; ok {
+					pattern = strings.Replace(pattern, subName[0], subPattern, 1)
+				}
+			}
+			p.patterns[name] = pattern
+		}
+	}
+
+	// check if pattern contains modifiers. Parse them out if it does.
+	for name, pattern := range p.patterns {
+		if modifierRe.MatchString(pattern) {
+			// this pattern has modifiers, so parse out the modifiers
+			pattern, err = p.parseTypedCaptures(name, pattern)
+			if err != nil {
+				return err
+			}
+			p.patterns[name] = pattern
+		}
+	}
+
+	return p.g.AddPatternsFromMap(p.patterns)
+}
+
+// parseTypedCaptures parses the capture modifiers, and then deletes the
+// modifier from the line so that it is a valid "grok" pattern again.
+//   ie,
+//     %{NUMBER:bytes:int}      => %{NUMBER:bytes}      (stores %{NUMBER}->bytes->int)
+//     %{IPORHOST:clientip:tag} => %{IPORHOST:clientip} (stores %{IPORHOST}->clientip->tag)
+func (p *Parser) parseTypedCaptures(name, pattern string) (string, error) {
+	matches := modifierRe.FindAllStringSubmatch(pattern, -1)
+
+	// grab the name of the capture pattern
+	patternName := "%{" + name + "}"
+	// create type map for this pattern
+	p.typeMap[patternName] = make(map[string]string)
+	p.tsMap[patternName] = make(map[string]string)
+
+	// boolean to verify that each pattern only has a single ts- data type.
+	hasTimestamp := false
+	for _, match := range matches {
+		// regex capture 1 is the name of the capture
+		// regex capture 2 is the modifier of the capture
+		if strings.HasPrefix(match[2], "ts") {
+			if hasTimestamp {
+				return pattern, fmt.Errorf("logparser pattern compile error: "+
+					"Each pattern is allowed only one named "+
+					"timestamp data type. pattern: %s", pattern)
+			}
+			if layout, ok := timeLayouts[match[2]]; ok {
+				// built-in time format
+				p.tsMap[patternName][match[1]] = layout
+			} else {
+				// custom time format
+				p.tsMap[patternName][match[1]] = strings.TrimSuffix(strings.TrimPrefix(match[2], `ts-"`), `"`)
+			}
+			hasTimestamp = true
+		} else {
+			p.typeMap[patternName][match[1]] = match[2]
+		}
+
+		// the modifier is not a valid part of a "grok" pattern, so remove it
+		// from the pattern.
+		pattern = strings.Replace(pattern, ":"+match[2]+"}", "}", 1)
+	}
+
+	return pattern, nil
+}
+
+// tsModder is a struct for incrementing identical timestamps of log lines
+// so that we don't push identical metrics that will get overwritten.
+type tsModder struct {
+	dupe     time.Time
+	last     time.Time
+	incr     time.Duration
+	incrn    time.Duration
+	rollover time.Duration
+}
+
+// tsMod increments the given timestamp one unit more from the previous
+// duplicate timestamp.
+// the increment unit is determined as the next smallest time unit below the
+// most significant time unit of ts.
+//   ie, if the input is at ms precision, it will increment it 1µs.
+func (t *tsModder) tsMod(ts time.Time) time.Time {
+	defer func() { t.last = ts }()
+	// don't mod the time if we don't need to
+	if t.last.IsZero() || ts.IsZero() {
+		t.incrn = 0
+		t.rollover = 0
+		return ts
+	}
+	if !ts.Equal(t.last) && !ts.Equal(t.dupe) {
+		t.incr = 0
+		t.incrn = 0
+		t.rollover = 0
+		return ts
+	}
+
+	if ts.Equal(t.last) {
+		t.dupe = ts
+	}
+
+	if ts.Equal(t.dupe) && t.incr == time.Duration(0) {
+		tsNano := ts.UnixNano()
+
+		d := int64(10)
+		counter := 1
+		for {
+			a := tsNano % d
+			if a > 0 {
+				break
+			}
+			d = d * 10
+			counter++
+		}
+
+		switch {
+		case counter <= 6:
+			t.incr = time.Nanosecond
+		case counter <= 9:
+			t.incr = time.Microsecond
+		case counter > 9:
+			t.incr = time.Millisecond
+		}
+	}
+
+	t.incrn++
+	if t.incrn == 999 && t.incr > time.Nanosecond {
+		t.rollover = t.incr * t.incrn
+		t.incrn = 1
+		t.incr = t.incr / 1000
+		if t.incr < time.Nanosecond {
+			t.incr = time.Nanosecond
+		}
+	}
+	return ts.Add(t.incr*t.incrn + t.rollover)
+}

plugins/parsers/grok/parser_test.go

@@ -0,0 +1,19 @@
+package grok
+
+import (
+	"log"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+)
+
+func TestGrokParse(t *testing.T) {
+	parser := Parser{
+		Measurement: "t_met",
+		Patterns:    []string{"%{COMMON_LOG_FORMAT}"},
+	}
+	parser.Compile()
+	metrics, err := parser.Parse([]byte(`127.0.0.1 user-identifier frank [10/Oct/2000:13:55:36 -0700] "GET /apache_pb.gif HTTP/1.0" 200 2326`))
+	log.Printf("metric_tags: %v, metric_fields: %v", metrics[0].Tags(), metrics[0].Fields())
+	assert.NoError(t, err)
+}

plugins/parsers/registry.go

@@ -8,6 +8,7 @@ import (
 	"github.com/influxdata/telegraf/plugins/parsers/collectd"
 	"github.com/influxdata/telegraf/plugins/parsers/dropwizard"
 	"github.com/influxdata/telegraf/plugins/parsers/graphite"
+	"github.com/influxdata/telegraf/plugins/parsers/grok"
 	"github.com/influxdata/telegraf/plugins/parsers/influx"
 	"github.com/influxdata/telegraf/plugins/parsers/json"
 	"github.com/influxdata/telegraf/plugins/parsers/nagios"
@@ -87,6 +88,13 @@ type Config struct {
 	// an optional map containing tag names as keys and json paths to retrieve the tag values from as values
 	// used if TagsPath is empty or doesn't return any tags
 	DropwizardTagPathsMap map[string]string
+
+	//grok patterns
+	Patterns           []string
+	NamedPatterns      []string
+	CustomPatterns     string
+	CustomPatternFiles []string
+	TimeZone           string
 }
 
 // NewParser returns a Parser interface based on the given config.
@@ -120,12 +128,39 @@ func NewParser(config *Config) (Parser, error) {
 			config.DefaultTags,
 			config.Separator,
 			config.Templates)
+
+	case "grok":
+		parser, err = NewGrokParser(
+			config.MetricName,
+			config.Patterns,
+			config.NamedPatterns,
+			config.CustomPatterns,
+			config.CustomPatternFiles,
+			config.TimeZone)
 	default:
 		err = fmt.Errorf("Invalid data format: %s", config.DataFormat)
 	}
 	return parser, err
 }
 
+func NewGrokParser(metricName string,
+	patterns []string,
+	nPatterns []string,
+	cPatterns string,
+	cPatternFiles []string, tZone string) (Parser, error) {
+	parser := grok.Parser{
+		Measurement:        metricName,
+		Patterns:           patterns,
+		NamedPatterns:      nPatterns,
+		CustomPatterns:     cPatterns,
+		CustomPatternFiles: cPatternFiles,
+		Timezone:           tZone,
+	}
+
+	parser.Compile()
+	return &parser, nil
+}
+
 func NewJSONParser(
 	metricName string,
 	tagKeys []string,

telegraf.conf

@@ -0,0 +1,104 @@
+
+# Global tags can be specified here in key="value" format.
+[global_tags]
+  # dc = "us-east-1" # will tag all metrics with dc=us-east-1
+  # rack = "1a"
+  ## Environment variables can be used as tags, and throughout the config file
+  # user = "$USER"
+
+
+# Configuration for telegraf agent
+[agent]
+  ## Default data collection interval for all inputs
+  interval = "10s"
+  ## Rounds collection interval to 'interval'
+  ## ie, if interval="10s" then always collect on :00, :10, :20, etc.
+  round_interval = true
+
+  ## Telegraf will send metrics to outputs in batches of at most
+  ## metric_batch_size metrics.
+  ## This controls the size of writes that Telegraf sends to output plugins.
+  metric_batch_size = 1000
+
+  ## For failed writes, telegraf will cache metric_buffer_limit metrics for each
+  ## output, and will flush this buffer on a successful write. Oldest metrics
+  ## are dropped first when this buffer fills.
+  ## This buffer only fills when writes fail to output plugin(s).
+  metric_buffer_limit = 10000
+
+  ## Collection jitter is used to jitter the collection by a random amount.
+  ## Each plugin will sleep for a random time within jitter before collecting.
+  ## This can be used to avoid many plugins querying things like sysfs at the
+  ## same time, which can have a measurable effect on the system.
+  collection_jitter = "0s"
+
+  ## Default flushing interval for all outputs. You shouldn't set this below
+  ## interval. Maximum flush_interval will be flush_interval + flush_jitter
+  flush_interval = "10s"
+  ## Jitter the flush interval by a random amount. This is primarily to avoid
+  ## large write spikes for users running a large number of telegraf instances.
+  ## ie, a jitter of 5s and interval 10s means flushes will happen every 10-15s
+  flush_jitter = "0s"
+
+  ## By default or when set to "0s", precision will be set to the same
+  ## timestamp order as the collection interval, with the maximum being 1s.
+  ##   ie, when interval = "10s", precision will be "1s"
+  ##       when interval = "250ms", precision will be "1ms"
+  ## Precision will NOT be used for service inputs. It is up to each individual
+  ## service input to set the timestamp at the appropriate precision.
+  ## Valid time units are "ns", "us" (or "µs"), "ms", "s".
+  precision = ""
+
+  ## Logging configuration:
+  ## Run telegraf with debug log messages.
+  debug = false
+  ## Run telegraf in quiet mode (error log messages only).
+  quiet = false
+  ## Specify the log file name. The empty string means to log to stderr.
+  logfile = ""
+
+  ## Override default hostname, if empty use os.Hostname()
+  hostname = ""
+  ## If set to true, do no set the "host" tag in the telegraf agent.
+  omit_hostname = false
+
+# # reload and gather from file[s] on telegraf's interval
+[[inputs.reader]]
+# ## These accept standard unix glob matching rules, but with the addition of
+# ## ** as a "super asterisk". ie:
+# ##   /var/log/**.log     -> recursively find all .log files in /var/log
+# ##   /var/log/*/*.log    -> find all .log files with a parent dir in /var/log
+# ##   /var/log/apache.log -> only tail the apache log file
+files = ["/Users/maxu/go/src/github.com/influxdata/telegraf/plugins/inputs/logparser/grok/testdata/**.log"]
+#
+# ## The dataformat to be read from files
+# ## Each data format has its own unique set of configuration options, read
+# ## more about them here:
+# ## https://github.com/influxdata/telegraf/blob/master/docs/DATA_FORMATS_INPUT.md
+data_format = "grok"
+#
+
+patterns = ["%{TEST_LOG_B}","%{TEST_LOG_A}"]
+#
+#   ## Name of the outputted measurement name.
+name_override = "grok_reader"
+#
+#   ## Full path(s) to custom pattern files.
+custom_pattern_files = ["/Users/maxu/go/src/github.com/influxdata/telegraf/plugins/inputs/logparser/grok/testdata/test-patterns"]
+#
+#   ## Custom patterns can also be defined here. Put one pattern per line.
+#   custom_patterns = '''
+#   '''
+#
+#   ## Timezone allows you to provide an override for timestamps that
+#   ## don't already include an offset
+#   ## e.g. 04/06/2016 12:41:45 data one two 5.43µs
+#   ##
+#   ## Default: "" which renders UTC
+#   ## Options are as follows:
+#   ##   1. Local             -- interpret based on machine localtime
+#   ##   2. "Canada/Eastern"  -- Unix TZ values like those found in https://en.wikipedia.org/wiki/List_of_tz_database_time_zones
+#   ##   3. UTC               -- or blank/unspecified, will return timestamp in UTC
+#   timezone = "Canada/Eastern"
+
+