Compare commits
11 Commits
plugin/rea
...
plugin/rea
| Author | SHA1 | Date | |
|---|---|---|---|
|
|
e4b6f236b6 | ||
|
|
c6087abf09 | ||
|
|
cafa95e536 | ||
|
|
8a9da28734 | ||
|
|
67db14332f | ||
|
|
bfc13a744b | ||
|
|
8063b38b2d | ||
|
|
04f09d65bf | ||
|
|
892c95aa6e | ||
|
|
aa750ec2b8 | ||
|
|
1be2a8eeaf |
12
Makefile
12
Makefile
@@ -54,11 +54,11 @@ fmtcheck:
|
||||
@echo '[INFO] done.'
|
||||
|
||||
test-windows:
|
||||
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/...
|
||||
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/...
|
||||
|
||||
# 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
|
||||
|
||||
@@ -11,7 +11,6 @@ Telegraf is able to parse the following input data formats into metrics:
|
||||
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/),
|
||||
are a combination of four basic parts:
|
||||
@@ -656,34 +655,103 @@ For more information about the dropwizard json format see
|
||||
```
|
||||
|
||||
#### Grok
|
||||
Parse logstash-style "grok" patterns:
|
||||
Parse logstash-style "grok" patterns. Patterns can be added to patterns, or custom patterns read from custom_pattern_files.
|
||||
|
||||
# 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}
|
||||
|
||||
#### Grok Configuration:
|
||||
```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}"]
|
||||
[[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)
|
||||
grok_patterns = ["%{COMBINED_LOG_FORMAT}"]
|
||||
|
||||
## Name of the outputted measurement name.
|
||||
name_override = "apache_access_log"
|
||||
## Name of the outputted measurement name.
|
||||
grok_name_override = "apache_access_log"
|
||||
|
||||
## Full path(s) to custom pattern files.
|
||||
custom_pattern_files = []
|
||||
## Full path(s) to custom pattern files.
|
||||
grok_custom_pattern_files = []
|
||||
|
||||
## Custom patterns can also be defined here. Put one pattern per line.
|
||||
custom_patterns = '''
|
||||
## Custom patterns can also be defined here. Put one pattern per line.
|
||||
grok_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"
|
||||
```
|
||||
## 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
|
||||
grok_timezone = "Canada/Eastern"
|
||||
```
|
||||
@@ -211,12 +211,16 @@ 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
|
||||
@@ -224,20 +228,24 @@ 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"
|
||||
@@ -245,6 +253,7 @@ 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"
|
||||
@@ -253,6 +262,7 @@ 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
|
||||
@@ -260,34 +270,41 @@ 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 #
|
||||
###############################################################################
|
||||
@@ -1322,7 +1339,7 @@ func buildParser(name string, tbl *ast.Table) (parsers.Parser, error) {
|
||||
}
|
||||
|
||||
//for grok data_format
|
||||
if node, ok := tbl.Fields["named_patterns"]; ok {
|
||||
if node, ok := tbl.Fields["grok_named_patterns"]; ok {
|
||||
if kv, ok := node.(*ast.KeyValue); ok {
|
||||
if ary, ok := kv.Value.(*ast.Array); ok {
|
||||
for _, elem := range ary.Value {
|
||||
@@ -1334,7 +1351,7 @@ func buildParser(name string, tbl *ast.Table) (parsers.Parser, error) {
|
||||
}
|
||||
}
|
||||
|
||||
if node, ok := tbl.Fields["patterns"]; ok {
|
||||
if node, ok := tbl.Fields["grok_patterns"]; ok {
|
||||
if kv, ok := node.(*ast.KeyValue); ok {
|
||||
if ary, ok := kv.Value.(*ast.Array); ok {
|
||||
for _, elem := range ary.Value {
|
||||
@@ -1346,7 +1363,7 @@ func buildParser(name string, tbl *ast.Table) (parsers.Parser, error) {
|
||||
}
|
||||
}
|
||||
|
||||
if node, ok := tbl.Fields["custom_patterns"]; ok {
|
||||
if node, ok := tbl.Fields["grok_custom_patterns"]; ok {
|
||||
if kv, ok := node.(*ast.KeyValue); ok {
|
||||
if str, ok := kv.Value.(*ast.String); ok {
|
||||
c.CustomPatterns = str.Value
|
||||
@@ -1354,7 +1371,7 @@ func buildParser(name string, tbl *ast.Table) (parsers.Parser, error) {
|
||||
}
|
||||
}
|
||||
|
||||
if node, ok := tbl.Fields["custom_pattern_files"]; ok {
|
||||
if node, ok := tbl.Fields["grok_custom_pattern_files"]; ok {
|
||||
if kv, ok := node.(*ast.KeyValue); ok {
|
||||
if ary, ok := kv.Value.(*ast.Array); ok {
|
||||
for _, elem := range ary.Value {
|
||||
@@ -1366,7 +1383,7 @@ func buildParser(name string, tbl *ast.Table) (parsers.Parser, error) {
|
||||
}
|
||||
}
|
||||
|
||||
if node, ok := tbl.Fields["timezone"]; ok {
|
||||
if node, ok := tbl.Fields["grok_timezone"]; ok {
|
||||
if kv, ok := node.(*ast.KeyValue); ok {
|
||||
if str, ok := kv.Value.(*ast.String); ok {
|
||||
c.TimeZone = str.Value
|
||||
@@ -1389,11 +1406,11 @@ 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")
|
||||
delete(tbl.Fields, "grok_named_patterns")
|
||||
delete(tbl.Fields, "grok_patterns")
|
||||
delete(tbl.Fields, "grok_custom_patterns")
|
||||
delete(tbl.Fields, "grok_custom_pattern_files")
|
||||
delete(tbl.Fields, "grok_timezone")
|
||||
|
||||
return parsers.NewParser(c)
|
||||
}
|
||||
|
||||
@@ -1,511 +0,0 @@
|
||||
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:
|
||||
// Replace commas with dot character
|
||||
v = strings.Replace(v, ",", ".", -1)
|
||||
|
||||
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) 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)
|
||||
}
|
||||
File diff suppressed because it is too large
Load Diff
@@ -1,78 +0,0 @@
|
||||
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}
|
||||
`
|
||||
@@ -14,9 +14,8 @@ import (
|
||||
"github.com/influxdata/telegraf"
|
||||
"github.com/influxdata/telegraf/internal/globpath"
|
||||
"github.com/influxdata/telegraf/plugins/inputs"
|
||||
|
||||
"github.com/influxdata/telegraf/plugins/parsers"
|
||||
// Parsers
|
||||
"github.com/influxdata/telegraf/plugins/inputs/logparser/grok"
|
||||
)
|
||||
|
||||
const (
|
||||
@@ -36,9 +35,10 @@ type logEntry struct {
|
||||
|
||||
// LogParserPlugin is the primary struct to implement the interface for logparser plugin
|
||||
type LogParserPlugin struct {
|
||||
Files []string
|
||||
FromBeginning bool
|
||||
WatchMethod string
|
||||
Files []string
|
||||
FromBeginning bool
|
||||
WatchMethod string
|
||||
MeasurementName string `toml:"measurement"`
|
||||
|
||||
tailers map[string]*tail.Tail
|
||||
lines chan logEntry
|
||||
@@ -49,7 +49,13 @@ type LogParserPlugin struct {
|
||||
|
||||
sync.Mutex
|
||||
|
||||
GrokParser *grok.Parser `toml:"grok"`
|
||||
GrokParser parsers.Parser `toml:"grok"`
|
||||
|
||||
Patterns []string
|
||||
NamedPatterns []string
|
||||
CustomPatterns string
|
||||
CustomPatternFiles []string
|
||||
TimeZone string
|
||||
}
|
||||
|
||||
const sampleConfig = `
|
||||
@@ -132,6 +138,21 @@ func (l *LogParserPlugin) Start(acc telegraf.Accumulator) error {
|
||||
|
||||
// Looks for fields which implement LogParser interface
|
||||
l.parsers = []LogParser{}
|
||||
config := &parsers.Config{
|
||||
Patterns: l.Patterns,
|
||||
NamedPatterns: l.NamedPatterns,
|
||||
CustomPatterns: l.CustomPatterns,
|
||||
CustomPatternFiles: l.CustomPatternFiles,
|
||||
TimeZone: l.TimeZone,
|
||||
DataFormat: "grok",
|
||||
}
|
||||
|
||||
var err error
|
||||
l.GrokParser, err = parsers.NewParser(config)
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
s := reflect.ValueOf(l).Elem()
|
||||
for i := 0; i < s.NumField(); i++ {
|
||||
f := s.Field(i)
|
||||
@@ -152,13 +173,6 @@ func (l *LogParserPlugin) Start(acc telegraf.Accumulator) error {
|
||||
return fmt.Errorf("logparser input plugin: no parser defined")
|
||||
}
|
||||
|
||||
// compile log parser patterns:
|
||||
for _, parser := range l.parsers {
|
||||
if err := parser.Compile(); err != nil {
|
||||
return err
|
||||
}
|
||||
}
|
||||
|
||||
l.wg.Add(1)
|
||||
go l.parser()
|
||||
|
||||
@@ -247,8 +261,8 @@ func (l *LogParserPlugin) receiver(tailer *tail.Tail) {
|
||||
}
|
||||
}
|
||||
|
||||
// parser is launched as a goroutine to watch the l.lines channel.
|
||||
// when a line is available, parser parses it and adds the metric(s) to the
|
||||
// parse is launched as a goroutine to watch the l.lines channel.
|
||||
// when a line is available, parse parses it and adds the metric(s) to the
|
||||
// accumulator.
|
||||
func (l *LogParserPlugin) parser() {
|
||||
defer l.wg.Done()
|
||||
@@ -265,18 +279,17 @@ func (l *LogParserPlugin) parser() {
|
||||
continue
|
||||
}
|
||||
}
|
||||
for _, parser := range l.parsers {
|
||||
m, err = parser.ParseLine(entry.line)
|
||||
if err == nil {
|
||||
if m != nil {
|
||||
tags := m.Tags()
|
||||
tags["path"] = entry.path
|
||||
l.acc.AddFields(m.Name(), m.Fields(), tags, m.Time())
|
||||
}
|
||||
} else {
|
||||
log.Println("E! Error parsing log line: " + err.Error())
|
||||
m, err = l.GrokParser.ParseLine(entry.line)
|
||||
if err == nil {
|
||||
if m != nil {
|
||||
tags := m.Tags()
|
||||
tags["path"] = entry.path
|
||||
l.acc.AddFields(l.MeasurementName, m.Fields(), tags, m.Time())
|
||||
}
|
||||
} else {
|
||||
log.Println("E! Error parsing log line: " + err.Error())
|
||||
}
|
||||
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
@@ -2,6 +2,7 @@ package logparser
|
||||
|
||||
import (
|
||||
"io/ioutil"
|
||||
"log"
|
||||
"os"
|
||||
"runtime"
|
||||
"strings"
|
||||
@@ -9,8 +10,6 @@ import (
|
||||
|
||||
"github.com/influxdata/telegraf/testutil"
|
||||
|
||||
"github.com/influxdata/telegraf/plugins/inputs/logparser/grok"
|
||||
|
||||
"github.com/stretchr/testify/assert"
|
||||
)
|
||||
|
||||
@@ -26,15 +25,12 @@ func TestStartNoParsers(t *testing.T) {
|
||||
|
||||
func TestGrokParseLogFilesNonExistPattern(t *testing.T) {
|
||||
thisdir := getCurrentDir()
|
||||
p := &grok.Parser{
|
||||
Patterns: []string{"%{FOOBAR}"},
|
||||
CustomPatternFiles: []string{thisdir + "grok/testdata/test-patterns"},
|
||||
}
|
||||
|
||||
logparser := &LogParserPlugin{
|
||||
FromBeginning: true,
|
||||
Files: []string{thisdir + "grok/testdata/*.log"},
|
||||
GrokParser: p,
|
||||
FromBeginning: true,
|
||||
Files: []string{thisdir + "grok/testdata/*.log"},
|
||||
Patterns: []string{"%{FOOBAR}"},
|
||||
CustomPatternFiles: []string{thisdir + "grok/testdata/test-patterns"},
|
||||
}
|
||||
|
||||
acc := testutil.Accumulator{}
|
||||
@@ -44,15 +40,13 @@ func TestGrokParseLogFilesNonExistPattern(t *testing.T) {
|
||||
|
||||
func TestGrokParseLogFiles(t *testing.T) {
|
||||
thisdir := getCurrentDir()
|
||||
p := &grok.Parser{
|
||||
Patterns: []string{"%{TEST_LOG_A}", "%{TEST_LOG_B}"},
|
||||
CustomPatternFiles: []string{thisdir + "grok/testdata/test-patterns"},
|
||||
}
|
||||
|
||||
logparser := &LogParserPlugin{
|
||||
FromBeginning: true,
|
||||
Files: []string{thisdir + "grok/testdata/*.log"},
|
||||
GrokParser: p,
|
||||
Patterns: []string{"%{TEST_LOG_A}", "%{TEST_LOG_B}"},
|
||||
CustomPatternFiles: []string{thisdir + "grok/testdata/test-patterns"},
|
||||
FromBeginning: true,
|
||||
Files: []string{thisdir + "grok/testdata/*.log"},
|
||||
MeasurementName: "logparser_grok",
|
||||
}
|
||||
|
||||
acc := testutil.Accumulator{}
|
||||
@@ -62,6 +56,7 @@ func TestGrokParseLogFiles(t *testing.T) {
|
||||
|
||||
logparser.Stop()
|
||||
|
||||
log.Printf("metric[0] %v, tags: %v, fields: %v", acc.Metrics[0].Measurement, acc.Metrics[0].Tags, acc.Metrics[0].Fields)
|
||||
acc.AssertContainsTaggedFields(t, "logparser_grok",
|
||||
map[string]interface{}{
|
||||
"clientip": "192.168.1.1",
|
||||
@@ -91,15 +86,13 @@ func TestGrokParseLogFilesAppearLater(t *testing.T) {
|
||||
assert.NoError(t, err)
|
||||
|
||||
thisdir := getCurrentDir()
|
||||
p := &grok.Parser{
|
||||
Patterns: []string{"%{TEST_LOG_A}", "%{TEST_LOG_B}"},
|
||||
CustomPatternFiles: []string{thisdir + "grok/testdata/test-patterns"},
|
||||
}
|
||||
|
||||
logparser := &LogParserPlugin{
|
||||
FromBeginning: true,
|
||||
Files: []string{emptydir + "/*.log"},
|
||||
GrokParser: p,
|
||||
FromBeginning: true,
|
||||
Files: []string{emptydir + "/*.log"},
|
||||
Patterns: []string{"%{TEST_LOG_A}", "%{TEST_LOG_B}"},
|
||||
CustomPatternFiles: []string{thisdir + "grok/testdata/test-patterns"},
|
||||
MeasurementName: "logparser_grok",
|
||||
}
|
||||
|
||||
acc := testutil.Accumulator{}
|
||||
@@ -130,16 +123,13 @@ func TestGrokParseLogFilesAppearLater(t *testing.T) {
|
||||
// pattern available for test_b.log
|
||||
func TestGrokParseLogFilesOneBad(t *testing.T) {
|
||||
thisdir := getCurrentDir()
|
||||
p := &grok.Parser{
|
||||
Patterns: []string{"%{TEST_LOG_A}", "%{TEST_LOG_BAD}"},
|
||||
CustomPatternFiles: []string{thisdir + "grok/testdata/test-patterns"},
|
||||
}
|
||||
assert.NoError(t, p.Compile())
|
||||
|
||||
logparser := &LogParserPlugin{
|
||||
FromBeginning: true,
|
||||
Files: []string{thisdir + "grok/testdata/test_a.log"},
|
||||
GrokParser: p,
|
||||
FromBeginning: true,
|
||||
Files: []string{thisdir + "grok/testdata/test_a.log"},
|
||||
Patterns: []string{"%{TEST_LOG_A}", "%{TEST_LOG_BAD}"},
|
||||
CustomPatternFiles: []string{thisdir + "grok/testdata/test-patterns"},
|
||||
MeasurementName: "logparser_grok",
|
||||
}
|
||||
|
||||
acc := testutil.Accumulator{}
|
||||
|
||||
@@ -1,12 +1,12 @@
|
||||
# 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
|
||||
The Reader Plugin updates a list of files every interval and parses the data inside.
|
||||
Files will always be read from the beginning.
|
||||
This plugin can parse any "data_format" formats.
|
||||
|
||||
### Configuration:
|
||||
```toml
|
||||
[[inputs.reader]]
|
||||
## Files to parse each interval.
|
||||
## These accept standard unix glob matching rules, but with the addition of
|
||||
## ** as a "super asterisk". ie:
|
||||
|
||||
@@ -11,4 +11,3 @@ services:
|
||||
- /telegraf
|
||||
- --config
|
||||
- /telegraf.conf
|
||||
|
||||
@@ -1,15 +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]
|
||||
}
|
||||
|
||||
"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]
|
||||
}
|
||||
|
||||
@@ -5,4 +5,3 @@
|
||||
|
||||
[[outputs.file]]
|
||||
files = ["stdout"]
|
||||
|
||||
@@ -1,8 +1,8 @@
|
||||
package reader
|
||||
|
||||
import (
|
||||
"fmt"
|
||||
"io/ioutil"
|
||||
"log"
|
||||
|
||||
"github.com/influxdata/telegraf"
|
||||
"github.com/influxdata/telegraf/internal/globpath"
|
||||
@@ -50,14 +50,8 @@ func (r *Reader) Gather(acc telegraf.Accumulator) error {
|
||||
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())
|
||||
for _, m := range metrics {
|
||||
acc.AddFields(m.Name(), m.Fields(), m.Tags(), m.Time())
|
||||
}
|
||||
}
|
||||
return nil
|
||||
@@ -67,13 +61,12 @@ func (r *Reader) SetParser(p parsers.Parser) {
|
||||
r.parser = p
|
||||
}
|
||||
|
||||
func (r *Reader) refreshFilePaths() {
|
||||
func (r *Reader) refreshFilePaths() error {
|
||||
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
|
||||
return fmt.Errorf("E! Error Glob: %v could not be compiled, %s", filepath, err)
|
||||
}
|
||||
files := g.Match()
|
||||
|
||||
@@ -83,13 +76,13 @@ func (r *Reader) refreshFilePaths() {
|
||||
}
|
||||
|
||||
r.Filenames = allFiles
|
||||
return nil
|
||||
}
|
||||
|
||||
//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 nil, fmt.Errorf("E! Error file: %v could not be read, %s", filename, err)
|
||||
}
|
||||
return r.parser.Parse(fileContents)
|
||||
|
||||
|
||||
@@ -71,7 +71,7 @@ type Parser struct {
|
||||
NamedPatterns []string
|
||||
CustomPatterns string
|
||||
CustomPatternFiles []string
|
||||
Measurement string
|
||||
MetricName string
|
||||
|
||||
// Timezone is an optional component to help render log dates to
|
||||
// your chosen zone.
|
||||
@@ -167,10 +167,6 @@ func (p *Parser) Compile() error {
|
||||
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)
|
||||
@@ -348,7 +344,7 @@ func (p *Parser) ParseLine(line string) (telegraf.Metric, error) {
|
||||
return nil, fmt.Errorf("logparser_grok: must have one or more fields")
|
||||
}
|
||||
|
||||
return metric.New(p.Measurement, tags, fields, p.tsModder.tsMod(timestamp))
|
||||
return metric.New(p.MetricName, tags, fields, p.tsModder.tsMod(timestamp))
|
||||
}
|
||||
|
||||
func (p *Parser) Parse(buf []byte) ([]telegraf.Metric, error) {
|
||||
|
||||
@@ -1,19 +1,906 @@
|
||||
package grok
|
||||
|
||||
import (
|
||||
"log"
|
||||
"testing"
|
||||
"time"
|
||||
|
||||
"github.com/stretchr/testify/assert"
|
||||
"github.com/stretchr/testify/require"
|
||||
)
|
||||
|
||||
func TestGrokParse(t *testing.T) {
|
||||
parser := Parser{
|
||||
Measurement: "t_met",
|
||||
Patterns: []string{"%{COMMON_LOG_FORMAT}"},
|
||||
MetricName: "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())
|
||||
_, 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`))
|
||||
assert.NoError(t, err)
|
||||
}
|
||||
|
||||
// Verify that patterns with a regex lookahead fail at compile time.
|
||||
func TestParsePatternsWithLookahead(t *testing.T) {
|
||||
p := &Parser{
|
||||
Patterns: []string{"%{MYLOG}"},
|
||||
CustomPatterns: `
|
||||
NOBOT ((?!bot|crawl).)*
|
||||
MYLOG %{NUMBER:num:int} %{NOBOT:client}
|
||||
`,
|
||||
}
|
||||
assert.NoError(t, p.Compile())
|
||||
|
||||
_, err := p.ParseLine(`1466004605359052000 bot`)
|
||||
assert.Error(t, err)
|
||||
}
|
||||
|
||||
func TestMeasurementName(t *testing.T) {
|
||||
p := &Parser{
|
||||
Patterns: []string{"%{COMMON_LOG_FORMAT}"},
|
||||
}
|
||||
assert.NoError(t, p.Compile())
|
||||
|
||||
// Parse an influxdb POST request
|
||||
m, err := p.ParseLine(`127.0.0.1 user-identifier frank [10/Oct/2000:13:55:36 -0700] "GET /apache_pb.gif HTTP/1.0" 200 2326`)
|
||||
require.NotNil(t, m)
|
||||
assert.NoError(t, err)
|
||||
assert.Equal(t,
|
||||
map[string]interface{}{
|
||||
"resp_bytes": int64(2326),
|
||||
"auth": "frank",
|
||||
"client_ip": "127.0.0.1",
|
||||
"http_version": float64(1.0),
|
||||
"ident": "user-identifier",
|
||||
"request": "/apache_pb.gif",
|
||||
},
|
||||
m.Fields())
|
||||
assert.Equal(t, map[string]string{"verb": "GET", "resp_code": "200"}, m.Tags())
|
||||
}
|
||||
|
||||
func TestCLF_IPv6(t *testing.T) {
|
||||
p := &Parser{
|
||||
Patterns: []string{"%{COMMON_LOG_FORMAT}"},
|
||||
}
|
||||
assert.NoError(t, p.Compile())
|
||||
|
||||
m, err := p.ParseLine(`2001:0db8:85a3:0000:0000:8a2e:0370:7334 user-identifier frank [10/Oct/2000:13:55:36 -0700] "GET /apache_pb.gif HTTP/1.0" 200 2326`)
|
||||
require.NotNil(t, m)
|
||||
assert.NoError(t, err)
|
||||
assert.Equal(t,
|
||||
map[string]interface{}{
|
||||
"resp_bytes": int64(2326),
|
||||
"auth": "frank",
|
||||
"client_ip": "2001:0db8:85a3:0000:0000:8a2e:0370:7334",
|
||||
"http_version": float64(1.0),
|
||||
"ident": "user-identifier",
|
||||
"request": "/apache_pb.gif",
|
||||
},
|
||||
m.Fields())
|
||||
assert.Equal(t, map[string]string{"verb": "GET", "resp_code": "200"}, m.Tags())
|
||||
|
||||
m, err = p.ParseLine(`::1 user-identifier frank [10/Oct/2000:13:55:36 -0700] "GET /apache_pb.gif HTTP/1.0" 200 2326`)
|
||||
require.NotNil(t, m)
|
||||
assert.NoError(t, err)
|
||||
assert.Equal(t,
|
||||
map[string]interface{}{
|
||||
"resp_bytes": int64(2326),
|
||||
"auth": "frank",
|
||||
"client_ip": "::1",
|
||||
"http_version": float64(1.0),
|
||||
"ident": "user-identifier",
|
||||
"request": "/apache_pb.gif",
|
||||
},
|
||||
m.Fields())
|
||||
assert.Equal(t, map[string]string{"verb": "GET", "resp_code": "200"}, m.Tags())
|
||||
}
|
||||
|
||||
func TestCustomInfluxdbHttpd(t *testing.T) {
|
||||
p := &Parser{
|
||||
Patterns: []string{`\[httpd\] %{COMBINED_LOG_FORMAT} %{UUID:uuid:drop} %{NUMBER:response_time_us:int}`},
|
||||
}
|
||||
assert.NoError(t, p.Compile())
|
||||
|
||||
// Parse an influxdb POST request
|
||||
m, err := p.ParseLine(`[httpd] ::1 - - [14/Jun/2016:11:33:29 +0100] "POST /write?consistency=any&db=telegraf&precision=ns&rp= HTTP/1.1" 204 0 "-" "InfluxDBClient" 6f61bc44-321b-11e6-8050-000000000000 2513`)
|
||||
require.NotNil(t, m)
|
||||
assert.NoError(t, err)
|
||||
assert.Equal(t,
|
||||
map[string]interface{}{
|
||||
"resp_bytes": int64(0),
|
||||
"auth": "-",
|
||||
"client_ip": "::1",
|
||||
"http_version": float64(1.1),
|
||||
"ident": "-",
|
||||
"referrer": "-",
|
||||
"request": "/write?consistency=any&db=telegraf&precision=ns&rp=",
|
||||
"response_time_us": int64(2513),
|
||||
"agent": "InfluxDBClient",
|
||||
},
|
||||
m.Fields())
|
||||
assert.Equal(t, map[string]string{"verb": "POST", "resp_code": "204"}, m.Tags())
|
||||
|
||||
// Parse an influxdb GET request
|
||||
m, err = p.ParseLine(`[httpd] ::1 - - [14/Jun/2016:12:10:02 +0100] "GET /query?db=telegraf&q=SELECT+bytes%2Cresponse_time_us+FROM+logparser_grok+WHERE+http_method+%3D+%27GET%27+AND+response_time_us+%3E+0+AND+time+%3E+now%28%29+-+1h HTTP/1.1" 200 578 "http://localhost:8083/" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/51.0.2704.84 Safari/537.36" 8a3806f1-3220-11e6-8006-000000000000 988`)
|
||||
require.NotNil(t, m)
|
||||
assert.NoError(t, err)
|
||||
assert.Equal(t,
|
||||
map[string]interface{}{
|
||||
"resp_bytes": int64(578),
|
||||
"auth": "-",
|
||||
"client_ip": "::1",
|
||||
"http_version": float64(1.1),
|
||||
"ident": "-",
|
||||
"referrer": "http://localhost:8083/",
|
||||
"request": "/query?db=telegraf&q=SELECT+bytes%2Cresponse_time_us+FROM+logparser_grok+WHERE+http_method+%3D+%27GET%27+AND+response_time_us+%3E+0+AND+time+%3E+now%28%29+-+1h",
|
||||
"response_time_us": int64(988),
|
||||
"agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/51.0.2704.84 Safari/537.36",
|
||||
},
|
||||
m.Fields())
|
||||
assert.Equal(t, map[string]string{"verb": "GET", "resp_code": "200"}, m.Tags())
|
||||
}
|
||||
|
||||
// common log format
|
||||
// 127.0.0.1 user-identifier frank [10/Oct/2000:13:55:36 -0700] "GET /apache_pb.gif HTTP/1.0" 200 2326
|
||||
func TestBuiltinCommonLogFormat(t *testing.T) {
|
||||
p := &Parser{
|
||||
Patterns: []string{"%{COMMON_LOG_FORMAT}"},
|
||||
}
|
||||
assert.NoError(t, p.Compile())
|
||||
|
||||
// Parse an influxdb POST request
|
||||
m, err := p.ParseLine(`127.0.0.1 user-identifier frank [10/Oct/2000:13:55:36 -0700] "GET /apache_pb.gif HTTP/1.0" 200 2326`)
|
||||
require.NotNil(t, m)
|
||||
assert.NoError(t, err)
|
||||
assert.Equal(t,
|
||||
map[string]interface{}{
|
||||
"resp_bytes": int64(2326),
|
||||
"auth": "frank",
|
||||
"client_ip": "127.0.0.1",
|
||||
"http_version": float64(1.0),
|
||||
"ident": "user-identifier",
|
||||
"request": "/apache_pb.gif",
|
||||
},
|
||||
m.Fields())
|
||||
assert.Equal(t, map[string]string{"verb": "GET", "resp_code": "200"}, m.Tags())
|
||||
}
|
||||
|
||||
// common log format
|
||||
// 127.0.0.1 user1234 frank1234 [10/Oct/2000:13:55:36 -0700] "GET /apache_pb.gif HTTP/1.0" 200 2326
|
||||
func TestBuiltinCommonLogFormatWithNumbers(t *testing.T) {
|
||||
p := &Parser{
|
||||
Patterns: []string{"%{COMMON_LOG_FORMAT}"},
|
||||
}
|
||||
assert.NoError(t, p.Compile())
|
||||
|
||||
// Parse an influxdb POST request
|
||||
m, err := p.ParseLine(`127.0.0.1 user1234 frank1234 [10/Oct/2000:13:55:36 -0700] "GET /apache_pb.gif HTTP/1.0" 200 2326`)
|
||||
require.NotNil(t, m)
|
||||
assert.NoError(t, err)
|
||||
assert.Equal(t,
|
||||
map[string]interface{}{
|
||||
"resp_bytes": int64(2326),
|
||||
"auth": "frank1234",
|
||||
"client_ip": "127.0.0.1",
|
||||
"http_version": float64(1.0),
|
||||
"ident": "user1234",
|
||||
"request": "/apache_pb.gif",
|
||||
},
|
||||
m.Fields())
|
||||
assert.Equal(t, map[string]string{"verb": "GET", "resp_code": "200"}, m.Tags())
|
||||
}
|
||||
|
||||
// combined log format
|
||||
// 127.0.0.1 user-identifier frank [10/Oct/2000:13:55:36 -0700] "GET /apache_pb.gif HTTP/1.0" 200 2326 "-" "Mozilla"
|
||||
func TestBuiltinCombinedLogFormat(t *testing.T) {
|
||||
p := &Parser{
|
||||
Patterns: []string{"%{COMBINED_LOG_FORMAT}"},
|
||||
}
|
||||
assert.NoError(t, p.Compile())
|
||||
|
||||
// Parse an influxdb POST request
|
||||
m, err := p.ParseLine(`127.0.0.1 user-identifier frank [10/Oct/2000:13:55:36 -0700] "GET /apache_pb.gif HTTP/1.0" 200 2326 "-" "Mozilla"`)
|
||||
require.NotNil(t, m)
|
||||
assert.NoError(t, err)
|
||||
assert.Equal(t,
|
||||
map[string]interface{}{
|
||||
"resp_bytes": int64(2326),
|
||||
"auth": "frank",
|
||||
"client_ip": "127.0.0.1",
|
||||
"http_version": float64(1.0),
|
||||
"ident": "user-identifier",
|
||||
"request": "/apache_pb.gif",
|
||||
"referrer": "-",
|
||||
"agent": "Mozilla",
|
||||
},
|
||||
m.Fields())
|
||||
assert.Equal(t, map[string]string{"verb": "GET", "resp_code": "200"}, m.Tags())
|
||||
}
|
||||
|
||||
func TestCompileStringAndParse(t *testing.T) {
|
||||
p := &Parser{
|
||||
Patterns: []string{"%{TEST_LOG_A}"},
|
||||
CustomPatterns: `
|
||||
DURATION %{NUMBER}[nuµm]?s
|
||||
RESPONSE_CODE %{NUMBER:response_code:tag}
|
||||
RESPONSE_TIME %{DURATION:response_time:duration}
|
||||
TEST_LOG_A %{NUMBER:myfloat:float} %{RESPONSE_CODE} %{IPORHOST:clientip} %{RESPONSE_TIME}
|
||||
`,
|
||||
}
|
||||
assert.NoError(t, p.Compile())
|
||||
|
||||
metricA, err := p.ParseLine(`1.25 200 192.168.1.1 5.432µs`)
|
||||
require.NotNil(t, metricA)
|
||||
assert.NoError(t, err)
|
||||
assert.Equal(t,
|
||||
map[string]interface{}{
|
||||
"clientip": "192.168.1.1",
|
||||
"myfloat": float64(1.25),
|
||||
"response_time": int64(5432),
|
||||
},
|
||||
metricA.Fields())
|
||||
assert.Equal(t, map[string]string{"response_code": "200"}, metricA.Tags())
|
||||
}
|
||||
|
||||
func TestCompileErrorsOnInvalidPattern(t *testing.T) {
|
||||
p := &Parser{
|
||||
Patterns: []string{"%{TEST_LOG_A}", "%{TEST_LOG_B}"},
|
||||
CustomPatterns: `
|
||||
DURATION %{NUMBER}[nuµm]?s
|
||||
RESPONSE_CODE %{NUMBER:response_code:tag}
|
||||
RESPONSE_TIME %{DURATION:response_time:duration}
|
||||
TEST_LOG_A %{NUMBER:myfloat:float} %{RESPONSE_CODE} %{IPORHOST:clientip} %{RESPONSE_TIME}
|
||||
`,
|
||||
}
|
||||
assert.Error(t, p.Compile())
|
||||
|
||||
metricA, _ := p.ParseLine(`1.25 200 192.168.1.1 5.432µs`)
|
||||
require.Nil(t, metricA)
|
||||
}
|
||||
|
||||
func TestParsePatternsWithoutCustom(t *testing.T) {
|
||||
p := &Parser{
|
||||
Patterns: []string{"%{POSINT:ts:ts-epochnano} response_time=%{POSINT:response_time:int} mymetric=%{NUMBER:metric:float}"},
|
||||
}
|
||||
assert.NoError(t, p.Compile())
|
||||
|
||||
metricA, err := p.ParseLine(`1466004605359052000 response_time=20821 mymetric=10890.645`)
|
||||
require.NotNil(t, metricA)
|
||||
assert.NoError(t, err)
|
||||
assert.Equal(t,
|
||||
map[string]interface{}{
|
||||
"response_time": int64(20821),
|
||||
"metric": float64(10890.645),
|
||||
},
|
||||
metricA.Fields())
|
||||
assert.Equal(t, map[string]string{}, metricA.Tags())
|
||||
assert.Equal(t, time.Unix(0, 1466004605359052000), metricA.Time())
|
||||
}
|
||||
|
||||
func TestParseEpochNano(t *testing.T) {
|
||||
p := &Parser{
|
||||
Patterns: []string{"%{MYAPP}"},
|
||||
CustomPatterns: `
|
||||
MYAPP %{POSINT:ts:ts-epochnano} response_time=%{POSINT:response_time:int} mymetric=%{NUMBER:metric:float}
|
||||
`,
|
||||
}
|
||||
assert.NoError(t, p.Compile())
|
||||
|
||||
metricA, err := p.ParseLine(`1466004605359052000 response_time=20821 mymetric=10890.645`)
|
||||
require.NotNil(t, metricA)
|
||||
assert.NoError(t, err)
|
||||
assert.Equal(t,
|
||||
map[string]interface{}{
|
||||
"response_time": int64(20821),
|
||||
"metric": float64(10890.645),
|
||||
},
|
||||
metricA.Fields())
|
||||
assert.Equal(t, map[string]string{}, metricA.Tags())
|
||||
assert.Equal(t, time.Unix(0, 1466004605359052000), metricA.Time())
|
||||
}
|
||||
|
||||
func TestParseEpoch(t *testing.T) {
|
||||
p := &Parser{
|
||||
Patterns: []string{"%{MYAPP}"},
|
||||
CustomPatterns: `
|
||||
MYAPP %{POSINT:ts:ts-epoch} response_time=%{POSINT:response_time:int} mymetric=%{NUMBER:metric:float}
|
||||
`,
|
||||
}
|
||||
assert.NoError(t, p.Compile())
|
||||
|
||||
metricA, err := p.ParseLine(`1466004605 response_time=20821 mymetric=10890.645`)
|
||||
require.NotNil(t, metricA)
|
||||
assert.NoError(t, err)
|
||||
assert.Equal(t,
|
||||
map[string]interface{}{
|
||||
"response_time": int64(20821),
|
||||
"metric": float64(10890.645),
|
||||
},
|
||||
metricA.Fields())
|
||||
assert.Equal(t, map[string]string{}, metricA.Tags())
|
||||
assert.Equal(t, time.Unix(1466004605, 0), metricA.Time())
|
||||
}
|
||||
|
||||
func TestParseEpochDecimal(t *testing.T) {
|
||||
var tests = []struct {
|
||||
name string
|
||||
line string
|
||||
noMatch bool
|
||||
err error
|
||||
tags map[string]string
|
||||
fields map[string]interface{}
|
||||
time time.Time
|
||||
}{
|
||||
{
|
||||
name: "ns precision",
|
||||
line: "1466004605.359052000 value=42",
|
||||
tags: map[string]string{},
|
||||
fields: map[string]interface{}{
|
||||
"value": int64(42),
|
||||
},
|
||||
time: time.Unix(0, 1466004605359052000),
|
||||
},
|
||||
{
|
||||
name: "ms precision",
|
||||
line: "1466004605.359 value=42",
|
||||
tags: map[string]string{},
|
||||
fields: map[string]interface{}{
|
||||
"value": int64(42),
|
||||
},
|
||||
time: time.Unix(0, 1466004605359000000),
|
||||
},
|
||||
{
|
||||
name: "second precision",
|
||||
line: "1466004605 value=42",
|
||||
tags: map[string]string{},
|
||||
fields: map[string]interface{}{
|
||||
"value": int64(42),
|
||||
},
|
||||
time: time.Unix(0, 1466004605000000000),
|
||||
},
|
||||
{
|
||||
name: "sub ns precision",
|
||||
line: "1466004605.123456789123 value=42",
|
||||
tags: map[string]string{},
|
||||
fields: map[string]interface{}{
|
||||
"value": int64(42),
|
||||
},
|
||||
time: time.Unix(0, 1466004605123456789),
|
||||
},
|
||||
}
|
||||
for _, tt := range tests {
|
||||
t.Run(tt.name, func(t *testing.T) {
|
||||
parser := &Parser{
|
||||
Patterns: []string{"%{NUMBER:ts:ts-epoch} value=%{NUMBER:value:int}"},
|
||||
}
|
||||
assert.NoError(t, parser.Compile())
|
||||
m, err := parser.ParseLine(tt.line)
|
||||
|
||||
if tt.noMatch {
|
||||
require.Nil(t, m)
|
||||
require.Nil(t, err)
|
||||
return
|
||||
}
|
||||
|
||||
require.Equal(t, tt.err, err)
|
||||
|
||||
require.NotNil(t, m)
|
||||
require.Equal(t, tt.tags, m.Tags())
|
||||
require.Equal(t, tt.fields, m.Fields())
|
||||
require.Equal(t, tt.time, m.Time())
|
||||
})
|
||||
}
|
||||
}
|
||||
|
||||
func TestParseEpochErrors(t *testing.T) {
|
||||
p := &Parser{
|
||||
Patterns: []string{"%{MYAPP}"},
|
||||
CustomPatterns: `
|
||||
MYAPP %{WORD:ts:ts-epoch} response_time=%{POSINT:response_time:int} mymetric=%{NUMBER:metric:float}
|
||||
`,
|
||||
}
|
||||
assert.NoError(t, p.Compile())
|
||||
|
||||
_, err := p.ParseLine(`foobar response_time=20821 mymetric=10890.645`)
|
||||
assert.NoError(t, err)
|
||||
|
||||
p = &Parser{
|
||||
Patterns: []string{"%{MYAPP}"},
|
||||
CustomPatterns: `
|
||||
MYAPP %{WORD:ts:ts-epochnano} response_time=%{POSINT:response_time:int} mymetric=%{NUMBER:metric:float}
|
||||
`,
|
||||
}
|
||||
assert.NoError(t, p.Compile())
|
||||
|
||||
_, err = p.ParseLine(`foobar response_time=20821 mymetric=10890.645`)
|
||||
assert.NoError(t, err)
|
||||
}
|
||||
|
||||
func TestParseGenericTimestamp(t *testing.T) {
|
||||
p := &Parser{
|
||||
Patterns: []string{`\[%{HTTPDATE:ts:ts}\] response_time=%{POSINT:response_time:int} mymetric=%{NUMBER:metric:float}`},
|
||||
}
|
||||
assert.NoError(t, p.Compile())
|
||||
|
||||
metricA, err := p.ParseLine(`[09/Jun/2016:03:37:03 +0000] response_time=20821 mymetric=10890.645`)
|
||||
require.NotNil(t, metricA)
|
||||
assert.NoError(t, err)
|
||||
assert.Equal(t,
|
||||
map[string]interface{}{
|
||||
"response_time": int64(20821),
|
||||
"metric": float64(10890.645),
|
||||
},
|
||||
metricA.Fields())
|
||||
assert.Equal(t, map[string]string{}, metricA.Tags())
|
||||
assert.Equal(t, time.Unix(1465443423, 0).UTC(), metricA.Time().UTC())
|
||||
|
||||
metricB, err := p.ParseLine(`[09/Jun/2016:03:37:04 +0000] response_time=20821 mymetric=10890.645`)
|
||||
require.NotNil(t, metricB)
|
||||
assert.NoError(t, err)
|
||||
assert.Equal(t,
|
||||
map[string]interface{}{
|
||||
"response_time": int64(20821),
|
||||
"metric": float64(10890.645),
|
||||
},
|
||||
metricB.Fields())
|
||||
assert.Equal(t, map[string]string{}, metricB.Tags())
|
||||
assert.Equal(t, time.Unix(1465443424, 0).UTC(), metricB.Time().UTC())
|
||||
}
|
||||
|
||||
func TestParseGenericTimestampNotFound(t *testing.T) {
|
||||
p := &Parser{
|
||||
Patterns: []string{`\[%{NOTSPACE:ts:ts}\] response_time=%{POSINT:response_time:int} mymetric=%{NUMBER:metric:float}`},
|
||||
}
|
||||
assert.NoError(t, p.Compile())
|
||||
|
||||
metricA, err := p.ParseLine(`[foobar] response_time=20821 mymetric=10890.645`)
|
||||
require.NotNil(t, metricA)
|
||||
assert.NoError(t, err)
|
||||
assert.Equal(t,
|
||||
map[string]interface{}{
|
||||
"response_time": int64(20821),
|
||||
"metric": float64(10890.645),
|
||||
},
|
||||
metricA.Fields())
|
||||
assert.Equal(t, map[string]string{}, metricA.Tags())
|
||||
}
|
||||
|
||||
func TestCompileFileAndParse(t *testing.T) {
|
||||
p := &Parser{
|
||||
Patterns: []string{"%{TEST_LOG_A}", "%{TEST_LOG_B}"},
|
||||
CustomPatternFiles: []string{"./testdata/test-patterns"},
|
||||
}
|
||||
assert.NoError(t, p.Compile())
|
||||
|
||||
metricA, err := p.ParseLine(`[04/Jun/2016:12:41:45 +0100] 1.25 200 192.168.1.1 5.432µs 101`)
|
||||
require.NotNil(t, metricA)
|
||||
assert.NoError(t, err)
|
||||
assert.Equal(t,
|
||||
map[string]interface{}{
|
||||
"clientip": "192.168.1.1",
|
||||
"myfloat": float64(1.25),
|
||||
"response_time": int64(5432),
|
||||
"myint": int64(101),
|
||||
},
|
||||
metricA.Fields())
|
||||
assert.Equal(t, map[string]string{"response_code": "200"}, metricA.Tags())
|
||||
assert.Equal(t,
|
||||
time.Date(2016, time.June, 4, 12, 41, 45, 0, time.FixedZone("foo", 60*60)).Nanosecond(),
|
||||
metricA.Time().Nanosecond())
|
||||
|
||||
metricB, err := p.ParseLine(`[04/06/2016--12:41:45] 1.25 mystring dropme nomodifier`)
|
||||
require.NotNil(t, metricB)
|
||||
assert.NoError(t, err)
|
||||
assert.Equal(t,
|
||||
map[string]interface{}{
|
||||
"myfloat": 1.25,
|
||||
"mystring": "mystring",
|
||||
"nomodifier": "nomodifier",
|
||||
},
|
||||
metricB.Fields())
|
||||
assert.Equal(t, map[string]string{}, metricB.Tags())
|
||||
assert.Equal(t,
|
||||
time.Date(2016, time.June, 4, 12, 41, 45, 0, time.FixedZone("foo", 60*60)).Nanosecond(),
|
||||
metricB.Time().Nanosecond())
|
||||
}
|
||||
|
||||
func TestCompileNoModifiersAndParse(t *testing.T) {
|
||||
p := &Parser{
|
||||
Patterns: []string{"%{TEST_LOG_C}"},
|
||||
CustomPatterns: `
|
||||
DURATION %{NUMBER}[nuµm]?s
|
||||
TEST_LOG_C %{NUMBER:myfloat} %{NUMBER} %{IPORHOST:clientip} %{DURATION:rt}
|
||||
`,
|
||||
}
|
||||
assert.NoError(t, p.Compile())
|
||||
|
||||
metricA, err := p.ParseLine(`1.25 200 192.168.1.1 5.432µs`)
|
||||
require.NotNil(t, metricA)
|
||||
assert.NoError(t, err)
|
||||
assert.Equal(t,
|
||||
map[string]interface{}{
|
||||
"clientip": "192.168.1.1",
|
||||
"myfloat": "1.25",
|
||||
"rt": "5.432µs",
|
||||
},
|
||||
metricA.Fields())
|
||||
assert.Equal(t, map[string]string{}, metricA.Tags())
|
||||
}
|
||||
|
||||
func TestCompileNoNamesAndParse(t *testing.T) {
|
||||
p := &Parser{
|
||||
Patterns: []string{"%{TEST_LOG_C}"},
|
||||
CustomPatterns: `
|
||||
DURATION %{NUMBER}[nuµm]?s
|
||||
TEST_LOG_C %{NUMBER} %{NUMBER} %{IPORHOST} %{DURATION}
|
||||
`,
|
||||
}
|
||||
assert.NoError(t, p.Compile())
|
||||
|
||||
metricA, err := p.ParseLine(`1.25 200 192.168.1.1 5.432µs`)
|
||||
require.Nil(t, metricA)
|
||||
assert.NoError(t, err)
|
||||
}
|
||||
|
||||
func TestParseNoMatch(t *testing.T) {
|
||||
p := &Parser{
|
||||
Patterns: []string{"%{TEST_LOG_A}", "%{TEST_LOG_B}"},
|
||||
CustomPatternFiles: []string{"./testdata/test-patterns"},
|
||||
}
|
||||
assert.NoError(t, p.Compile())
|
||||
|
||||
metricA, err := p.ParseLine(`[04/Jun/2016:12:41:45 +0100] notnumber 200 192.168.1.1 5.432µs 101`)
|
||||
assert.NoError(t, err)
|
||||
assert.Nil(t, metricA)
|
||||
}
|
||||
|
||||
func TestCompileErrors(t *testing.T) {
|
||||
// Compile fails because there are multiple timestamps:
|
||||
p := &Parser{
|
||||
Patterns: []string{"%{TEST_LOG_A}", "%{TEST_LOG_B}"},
|
||||
CustomPatterns: `
|
||||
TEST_LOG_A %{HTTPDATE:ts1:ts-httpd} %{HTTPDATE:ts2:ts-httpd} %{NUMBER:mynum:int}
|
||||
`,
|
||||
}
|
||||
assert.Error(t, p.Compile())
|
||||
|
||||
// Compile fails because file doesn't exist:
|
||||
p = &Parser{
|
||||
Patterns: []string{"%{TEST_LOG_A}", "%{TEST_LOG_B}"},
|
||||
CustomPatternFiles: []string{"/tmp/foo/bar/baz"},
|
||||
}
|
||||
assert.Error(t, p.Compile())
|
||||
}
|
||||
|
||||
func TestParseErrors(t *testing.T) {
|
||||
// Parse fails because the pattern doesn't exist
|
||||
p := &Parser{
|
||||
Patterns: []string{"%{TEST_LOG_B}"},
|
||||
CustomPatterns: `
|
||||
TEST_LOG_A %{HTTPDATE:ts:ts-httpd} %{WORD:myword:int} %{}
|
||||
`,
|
||||
}
|
||||
assert.Error(t, p.Compile())
|
||||
_, err := p.ParseLine(`[04/Jun/2016:12:41:45 +0100] notnumber 200 192.168.1.1 5.432µs 101`)
|
||||
assert.Error(t, err)
|
||||
|
||||
// Parse fails because myword is not an int
|
||||
p = &Parser{
|
||||
Patterns: []string{"%{TEST_LOG_A}"},
|
||||
CustomPatterns: `
|
||||
TEST_LOG_A %{HTTPDATE:ts:ts-httpd} %{WORD:myword:int}
|
||||
`,
|
||||
}
|
||||
assert.NoError(t, p.Compile())
|
||||
_, err = p.ParseLine(`04/Jun/2016:12:41:45 +0100 notnumber`)
|
||||
assert.Error(t, err)
|
||||
|
||||
// Parse fails because myword is not a float
|
||||
p = &Parser{
|
||||
Patterns: []string{"%{TEST_LOG_A}"},
|
||||
CustomPatterns: `
|
||||
TEST_LOG_A %{HTTPDATE:ts:ts-httpd} %{WORD:myword:float}
|
||||
`,
|
||||
}
|
||||
assert.NoError(t, p.Compile())
|
||||
_, err = p.ParseLine(`04/Jun/2016:12:41:45 +0100 notnumber`)
|
||||
assert.Error(t, err)
|
||||
|
||||
// Parse fails because myword is not a duration
|
||||
p = &Parser{
|
||||
Patterns: []string{"%{TEST_LOG_A}"},
|
||||
CustomPatterns: `
|
||||
TEST_LOG_A %{HTTPDATE:ts:ts-httpd} %{WORD:myword:duration}
|
||||
`,
|
||||
}
|
||||
assert.NoError(t, p.Compile())
|
||||
_, err = p.ParseLine(`04/Jun/2016:12:41:45 +0100 notnumber`)
|
||||
assert.Error(t, err)
|
||||
|
||||
// Parse fails because the time layout is wrong.
|
||||
p = &Parser{
|
||||
Patterns: []string{"%{TEST_LOG_A}"},
|
||||
CustomPatterns: `
|
||||
TEST_LOG_A %{HTTPDATE:ts:ts-unix} %{WORD:myword:duration}
|
||||
`,
|
||||
}
|
||||
assert.NoError(t, p.Compile())
|
||||
_, err = p.ParseLine(`04/Jun/2016:12:41:45 +0100 notnumber`)
|
||||
assert.Error(t, err)
|
||||
}
|
||||
|
||||
func TestTsModder(t *testing.T) {
|
||||
tsm := &tsModder{}
|
||||
|
||||
reftime := time.Date(2006, time.December, 1, 1, 1, 1, int(time.Millisecond), time.UTC)
|
||||
modt := tsm.tsMod(reftime)
|
||||
assert.Equal(t, reftime, modt)
|
||||
modt = tsm.tsMod(reftime)
|
||||
assert.Equal(t, reftime.Add(time.Microsecond*1), modt)
|
||||
modt = tsm.tsMod(reftime)
|
||||
assert.Equal(t, reftime.Add(time.Microsecond*2), modt)
|
||||
modt = tsm.tsMod(reftime)
|
||||
assert.Equal(t, reftime.Add(time.Microsecond*3), modt)
|
||||
|
||||
reftime = time.Date(2006, time.December, 1, 1, 1, 1, int(time.Microsecond), time.UTC)
|
||||
modt = tsm.tsMod(reftime)
|
||||
assert.Equal(t, reftime, modt)
|
||||
modt = tsm.tsMod(reftime)
|
||||
assert.Equal(t, reftime.Add(time.Nanosecond*1), modt)
|
||||
modt = tsm.tsMod(reftime)
|
||||
assert.Equal(t, reftime.Add(time.Nanosecond*2), modt)
|
||||
modt = tsm.tsMod(reftime)
|
||||
assert.Equal(t, reftime.Add(time.Nanosecond*3), modt)
|
||||
|
||||
reftime = time.Date(2006, time.December, 1, 1, 1, 1, int(time.Microsecond)*999, time.UTC)
|
||||
modt = tsm.tsMod(reftime)
|
||||
assert.Equal(t, reftime, modt)
|
||||
modt = tsm.tsMod(reftime)
|
||||
assert.Equal(t, reftime.Add(time.Nanosecond*1), modt)
|
||||
modt = tsm.tsMod(reftime)
|
||||
assert.Equal(t, reftime.Add(time.Nanosecond*2), modt)
|
||||
modt = tsm.tsMod(reftime)
|
||||
assert.Equal(t, reftime.Add(time.Nanosecond*3), modt)
|
||||
|
||||
reftime = time.Date(2006, time.December, 1, 1, 1, 1, 0, time.UTC)
|
||||
modt = tsm.tsMod(reftime)
|
||||
assert.Equal(t, reftime, modt)
|
||||
modt = tsm.tsMod(reftime)
|
||||
assert.Equal(t, reftime.Add(time.Millisecond*1), modt)
|
||||
modt = tsm.tsMod(reftime)
|
||||
assert.Equal(t, reftime.Add(time.Millisecond*2), modt)
|
||||
modt = tsm.tsMod(reftime)
|
||||
assert.Equal(t, reftime.Add(time.Millisecond*3), modt)
|
||||
|
||||
reftime = time.Time{}
|
||||
modt = tsm.tsMod(reftime)
|
||||
assert.Equal(t, reftime, modt)
|
||||
}
|
||||
|
||||
func TestTsModder_Rollover(t *testing.T) {
|
||||
tsm := &tsModder{}
|
||||
|
||||
reftime := time.Date(2006, time.December, 1, 1, 1, 1, int(time.Millisecond), time.UTC)
|
||||
modt := tsm.tsMod(reftime)
|
||||
for i := 1; i < 1000; i++ {
|
||||
modt = tsm.tsMod(reftime)
|
||||
}
|
||||
assert.Equal(t, reftime.Add(time.Microsecond*999+time.Nanosecond), modt)
|
||||
|
||||
reftime = time.Date(2006, time.December, 1, 1, 1, 1, int(time.Microsecond), time.UTC)
|
||||
modt = tsm.tsMod(reftime)
|
||||
for i := 1; i < 1001; i++ {
|
||||
modt = tsm.tsMod(reftime)
|
||||
}
|
||||
assert.Equal(t, reftime.Add(time.Nanosecond*1000), modt)
|
||||
}
|
||||
|
||||
func TestShortPatternRegression(t *testing.T) {
|
||||
p := &Parser{
|
||||
Patterns: []string{"%{TS_UNIX:timestamp:ts-unix} %{NUMBER:value:int}"},
|
||||
CustomPatterns: `
|
||||
TS_UNIX %{DAY} %{MONTH} %{MONTHDAY} %{HOUR}:%{MINUTE}:%{SECOND} %{TZ} %{YEAR}
|
||||
`,
|
||||
}
|
||||
require.NoError(t, p.Compile())
|
||||
|
||||
metric, err := p.ParseLine(`Wed Apr 12 13:10:34 PST 2017 42`)
|
||||
require.NoError(t, err)
|
||||
require.NotNil(t, metric)
|
||||
|
||||
require.Equal(t,
|
||||
map[string]interface{}{
|
||||
"value": int64(42),
|
||||
},
|
||||
metric.Fields())
|
||||
}
|
||||
|
||||
func TestTimezoneEmptyCompileFileAndParse(t *testing.T) {
|
||||
p := &Parser{
|
||||
Patterns: []string{"%{TEST_LOG_A}", "%{TEST_LOG_B}"},
|
||||
CustomPatternFiles: []string{"./testdata/test-patterns"},
|
||||
Timezone: "",
|
||||
}
|
||||
assert.NoError(t, p.Compile())
|
||||
|
||||
metricA, err := p.ParseLine(`[04/Jun/2016:12:41:45 +0100] 1.25 200 192.168.1.1 5.432µs 101`)
|
||||
require.NotNil(t, metricA)
|
||||
assert.NoError(t, err)
|
||||
assert.Equal(t,
|
||||
map[string]interface{}{
|
||||
"clientip": "192.168.1.1",
|
||||
"myfloat": float64(1.25),
|
||||
"response_time": int64(5432),
|
||||
"myint": int64(101),
|
||||
},
|
||||
metricA.Fields())
|
||||
assert.Equal(t, map[string]string{"response_code": "200"}, metricA.Tags())
|
||||
assert.Equal(t, int64(1465040505000000000), metricA.Time().UnixNano())
|
||||
|
||||
metricB, err := p.ParseLine(`[04/06/2016--12:41:45] 1.25 mystring dropme nomodifier`)
|
||||
require.NotNil(t, metricB)
|
||||
assert.NoError(t, err)
|
||||
assert.Equal(t,
|
||||
map[string]interface{}{
|
||||
"myfloat": 1.25,
|
||||
"mystring": "mystring",
|
||||
"nomodifier": "nomodifier",
|
||||
},
|
||||
metricB.Fields())
|
||||
assert.Equal(t, map[string]string{}, metricB.Tags())
|
||||
assert.Equal(t, int64(1465044105000000000), metricB.Time().UnixNano())
|
||||
}
|
||||
|
||||
func TestTimezoneMalformedCompileFileAndParse(t *testing.T) {
|
||||
p := &Parser{
|
||||
Patterns: []string{"%{TEST_LOG_A}", "%{TEST_LOG_B}"},
|
||||
CustomPatternFiles: []string{"./testdata/test-patterns"},
|
||||
Timezone: "Something/Weird",
|
||||
}
|
||||
assert.NoError(t, p.Compile())
|
||||
|
||||
metricA, err := p.ParseLine(`[04/Jun/2016:12:41:45 +0100] 1.25 200 192.168.1.1 5.432µs 101`)
|
||||
require.NotNil(t, metricA)
|
||||
assert.NoError(t, err)
|
||||
assert.Equal(t,
|
||||
map[string]interface{}{
|
||||
"clientip": "192.168.1.1",
|
||||
"myfloat": float64(1.25),
|
||||
"response_time": int64(5432),
|
||||
"myint": int64(101),
|
||||
},
|
||||
metricA.Fields())
|
||||
assert.Equal(t, map[string]string{"response_code": "200"}, metricA.Tags())
|
||||
assert.Equal(t, int64(1465040505000000000), metricA.Time().UnixNano())
|
||||
|
||||
metricB, err := p.ParseLine(`[04/06/2016--12:41:45] 1.25 mystring dropme nomodifier`)
|
||||
require.NotNil(t, metricB)
|
||||
assert.NoError(t, err)
|
||||
assert.Equal(t,
|
||||
map[string]interface{}{
|
||||
"myfloat": 1.25,
|
||||
"mystring": "mystring",
|
||||
"nomodifier": "nomodifier",
|
||||
},
|
||||
metricB.Fields())
|
||||
assert.Equal(t, map[string]string{}, metricB.Tags())
|
||||
assert.Equal(t, int64(1465044105000000000), metricB.Time().UnixNano())
|
||||
}
|
||||
|
||||
func TestTimezoneEuropeCompileFileAndParse(t *testing.T) {
|
||||
p := &Parser{
|
||||
Patterns: []string{"%{TEST_LOG_A}", "%{TEST_LOG_B}"},
|
||||
CustomPatternFiles: []string{"./testdata/test-patterns"},
|
||||
Timezone: "Europe/Berlin",
|
||||
}
|
||||
assert.NoError(t, p.Compile())
|
||||
|
||||
metricA, err := p.ParseLine(`[04/Jun/2016:12:41:45 +0100] 1.25 200 192.168.1.1 5.432µs 101`)
|
||||
require.NotNil(t, metricA)
|
||||
assert.NoError(t, err)
|
||||
assert.Equal(t,
|
||||
map[string]interface{}{
|
||||
"clientip": "192.168.1.1",
|
||||
"myfloat": float64(1.25),
|
||||
"response_time": int64(5432),
|
||||
"myint": int64(101),
|
||||
},
|
||||
metricA.Fields())
|
||||
assert.Equal(t, map[string]string{"response_code": "200"}, metricA.Tags())
|
||||
assert.Equal(t, int64(1465040505000000000), metricA.Time().UnixNano())
|
||||
|
||||
metricB, err := p.ParseLine(`[04/06/2016--12:41:45] 1.25 mystring dropme nomodifier`)
|
||||
require.NotNil(t, metricB)
|
||||
assert.NoError(t, err)
|
||||
assert.Equal(t,
|
||||
map[string]interface{}{
|
||||
"myfloat": 1.25,
|
||||
"mystring": "mystring",
|
||||
"nomodifier": "nomodifier",
|
||||
},
|
||||
metricB.Fields())
|
||||
assert.Equal(t, map[string]string{}, metricB.Tags())
|
||||
assert.Equal(t, int64(1465036905000000000), metricB.Time().UnixNano())
|
||||
}
|
||||
|
||||
func TestTimezoneAmericasCompileFileAndParse(t *testing.T) {
|
||||
p := &Parser{
|
||||
Patterns: []string{"%{TEST_LOG_A}", "%{TEST_LOG_B}"},
|
||||
CustomPatternFiles: []string{"./testdata/test-patterns"},
|
||||
Timezone: "Canada/Eastern",
|
||||
}
|
||||
assert.NoError(t, p.Compile())
|
||||
|
||||
metricA, err := p.ParseLine(`[04/Jun/2016:12:41:45 +0100] 1.25 200 192.168.1.1 5.432µs 101`)
|
||||
require.NotNil(t, metricA)
|
||||
assert.NoError(t, err)
|
||||
assert.Equal(t,
|
||||
map[string]interface{}{
|
||||
"clientip": "192.168.1.1",
|
||||
"myfloat": float64(1.25),
|
||||
"response_time": int64(5432),
|
||||
"myint": int64(101),
|
||||
},
|
||||
metricA.Fields())
|
||||
assert.Equal(t, map[string]string{"response_code": "200"}, metricA.Tags())
|
||||
assert.Equal(t, int64(1465040505000000000), metricA.Time().UnixNano())
|
||||
|
||||
metricB, err := p.ParseLine(`[04/06/2016--12:41:45] 1.25 mystring dropme nomodifier`)
|
||||
require.NotNil(t, metricB)
|
||||
assert.NoError(t, err)
|
||||
assert.Equal(t,
|
||||
map[string]interface{}{
|
||||
"myfloat": 1.25,
|
||||
"mystring": "mystring",
|
||||
"nomodifier": "nomodifier",
|
||||
},
|
||||
metricB.Fields())
|
||||
assert.Equal(t, map[string]string{}, metricB.Tags())
|
||||
assert.Equal(t, int64(1465058505000000000), metricB.Time().UnixNano())
|
||||
}
|
||||
|
||||
func TestTimezoneLocalCompileFileAndParse(t *testing.T) {
|
||||
p := &Parser{
|
||||
Patterns: []string{"%{TEST_LOG_A}", "%{TEST_LOG_B}"},
|
||||
CustomPatternFiles: []string{"./testdata/test-patterns"},
|
||||
Timezone: "Local",
|
||||
}
|
||||
assert.NoError(t, p.Compile())
|
||||
|
||||
metricA, err := p.ParseLine(`[04/Jun/2016:12:41:45 +0100] 1.25 200 192.168.1.1 5.432µs 101`)
|
||||
require.NotNil(t, metricA)
|
||||
assert.NoError(t, err)
|
||||
assert.Equal(t,
|
||||
map[string]interface{}{
|
||||
"clientip": "192.168.1.1",
|
||||
"myfloat": float64(1.25),
|
||||
"response_time": int64(5432),
|
||||
"myint": int64(101),
|
||||
},
|
||||
metricA.Fields())
|
||||
assert.Equal(t, map[string]string{"response_code": "200"}, metricA.Tags())
|
||||
assert.Equal(t, int64(1465040505000000000), metricA.Time().UnixNano())
|
||||
|
||||
metricB, err := p.ParseLine(`[04/06/2016--12:41:45] 1.25 mystring dropme nomodifier`)
|
||||
require.NotNil(t, metricB)
|
||||
assert.NoError(t, err)
|
||||
assert.Equal(t,
|
||||
map[string]interface{}{
|
||||
"myfloat": 1.25,
|
||||
"mystring": "mystring",
|
||||
"nomodifier": "nomodifier",
|
||||
},
|
||||
metricB.Fields())
|
||||
assert.Equal(t, map[string]string{}, metricB.Tags())
|
||||
assert.Equal(t, time.Date(2016, time.June, 4, 12, 41, 45, 0, time.Local).UnixNano(), metricB.Time().UnixNano())
|
||||
}
|
||||
|
||||
func TestNewlineInPatterns(t *testing.T) {
|
||||
p := &Parser{
|
||||
Patterns: []string{`
|
||||
%{SYSLOGTIMESTAMP:timestamp}
|
||||
`},
|
||||
}
|
||||
require.NoError(t, p.Compile())
|
||||
m, err := p.ParseLine("Apr 10 05:11:57")
|
||||
require.NoError(t, err)
|
||||
require.NotNil(t, m)
|
||||
}
|
||||
|
||||
BIN
plugins/parsers/grok/testdata/.DS_Store
vendored
Normal file
BIN
plugins/parsers/grok/testdata/.DS_Store
vendored
Normal file
Binary file not shown.
@@ -128,9 +128,8 @@ func NewParser(config *Config) (Parser, error) {
|
||||
config.DefaultTags,
|
||||
config.Separator,
|
||||
config.Templates)
|
||||
|
||||
case "grok":
|
||||
parser, err = NewGrokParser(
|
||||
parser, err = newGrokParser(
|
||||
config.MetricName,
|
||||
config.Patterns,
|
||||
config.NamedPatterns,
|
||||
@@ -143,13 +142,13 @@ func NewParser(config *Config) (Parser, error) {
|
||||
return parser, err
|
||||
}
|
||||
|
||||
func NewGrokParser(metricName string,
|
||||
func newGrokParser(metricName string,
|
||||
patterns []string,
|
||||
nPatterns []string,
|
||||
cPatterns string,
|
||||
cPatternFiles []string, tZone string) (Parser, error) {
|
||||
parser := grok.Parser{
|
||||
Measurement: metricName,
|
||||
MetricName: metricName,
|
||||
Patterns: patterns,
|
||||
NamedPatterns: nPatterns,
|
||||
CustomPatterns: cPatterns,
|
||||
@@ -157,8 +156,8 @@ func NewGrokParser(metricName string,
|
||||
Timezone: tZone,
|
||||
}
|
||||
|
||||
parser.Compile()
|
||||
return &parser, nil
|
||||
err := parser.Compile()
|
||||
return &parser, err
|
||||
}
|
||||
|
||||
func NewJSONParser(
|
||||
|
||||
104
telegraf.conf
104
telegraf.conf
@@ -1,104 +0,0 @@
|
||||
|
||||
# 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"
|
||||
|
||||
|
||||
Reference in New Issue
Block a user