With this remote storage adapter, Prometheus can use Splunk as a long-term store for time-series metrics.
A docker image for the splunk storage adapter is available on Docker Hub at kebe/ropee.
# You must edit the following command for your env.
$ docker run -d --name ropee -p 9970:9970 \
-e LISTEN_ADDR=0.0.0.0:9970 \
-e SPLUNK_METRICS_INDEX=metrics \
-e SPLUNK_METRICS_SOURCETYPE=DaoCloud_promu_metrics \
-e SPLUNK_HEC_TOKEN=asddsa1-12312312-3123-2 \
-e SPLUNK_HEC_URL=https://192.168.1.1:8088 \
-e SPLUNK_URL=https://192.168.1.1:8089 \
-e TIMEOUT=60 \
-e DEBUG=0 \
kebe/ropee:latest
Usage of ./ropee:
-debug
Debug mode.
-listen-addr string
Sopee listen addr. (default "127.0.0.1:9970")
-log-file-path string
Log files path. (default "/var/log")
-splunk-hec-token string
Splunk Http event collector token.
-splunk-hec-url string
Splunk Http event collector url. (default "https://127.0.0.1:8088")
-splunk-metrics-index string
Index name. (default "*")
-splunk-metrics-sourcetype string
The prometheus sourcetype name. (default "DaoCloud_promu_metrics")
-splunk-url string
Splunk Manage Url. (default "https://127.0.0.1:8089")
-timeout int
API timeout seconds. (default 60)
Please follow splunk docs.
props.conf
[DaoCloud_promu_metrics]
DATETIME_CONFIG = CURRENT
TRANSFORMS-prometheus_to_metric = prometheus_metric_name_value, prometheus_metric_dims
NO_BINARY_CHECK = true
description = Prometheus Metrics.
SHOULD_LINEMERGE = false
pulldown_type = 1
category = Metrics
transforms.conf
[prometheus_metric_name_value]
REGEX = ^([^\s{]+)({[^}]+})? ([-+]?[0-9]*\.?[0-9]+([eE][-+]?[0-9]+)?)
FORMAT = metric_name::$1 ::$2 _value::$3
WRITE_META = true
[prometheus_metric_dims]
REGEX = ([a-zA-Z_][a-zA-Z0-9_]*)="([^"]*)"[, ]*
FORMAT = $1::"$2"
REPEAT_MATCH = true
WRITE_META = true
Please follow splunk docs.
...
remote_read:
- url: "http://127.0.0.1:9970/read"
# for remote read, you should set the basic auth which belongs splunk's user.
remote_write:
- url: "http://127.0.0.1:9970/write"
go mod download
go run main.go