globocom/redis-healthy

It retrieves metrics, periodically, from Redis (or sentinel) and send them to Logstash

redis-healthy

It retrieves metrics periodically from Redis (or sentinel) (such as latency, connected_clients, instantaneous_ops_per_sec and others) and send them to Logstash.

Plotting the metrics with Grafana

Tests

make test

Metrics

{
   client_longest_output_list: 15,
   instantaneous_input_kbps: 0,
   sync_partial_err: 0,
   latency: 361,
   connected_clients: 398,
   blocked_clients: 0,
   keyspace_hits: 201980,
   client: 'app-redis',
   instantaneous_ops_per_sec: 1092,
   instantaneous_output_kbps: 504,
   sync_full: 0,
   keyspace_misses: 1093,
   mem_fragmentation_ratio: 0,
   rejected_connections: 0,
   sync_partial_ok: 0
}

Options

Variable	Mandatory	Description
PROJECT	Y	An identifier for the metrics. It'll be send as "client": PROJECT + "-redis"
PING_FREQUENCY		Frequency in seconds that the metrics are fetched. Default: 10
REDIS_HOST	Y	Redis host with port. If you're using redis sentinel, then REDIS_HOST will host the sentinel hosts separated by commas ("host:port,host:port")
REDIS_PWD		Redis password. Default: ""
REDIS_SENTINEL		Whether you're using sentinel or not. Default: ""
REDIS_MASTER_NAME		Redis sentinel master name. Default: ""
REDIS_LATENCY_THRESHOLD		Redis latency threshold in ms. When any command take longer than the threshold, then it sends data about latency. After PING_FREQUENCY has passed, it sends 0. Default: ""
REDIS_METRICS_TO_WATCH		The fields you want to keep track from the output of the command "info". Default: "client_longest_output_list,connected_clients,blocked_clients,rejected_connections,instantaneous_input_kbps,instantaneous_output_kbps,instantaneous_ops_per_sec,keyspace_hits,keyspace_misses,mem_fragmentation_ratio,sync_full,sync_partial_ok,sync_partial_err"
LOGSTASH_HOST	Y	Logstash host
LOGSTASH_PORT	Y	Logstash port
LOGSTASH_PROTOCOL		The transport protocol used by logstash. Default: "udp"

Usage

REDIS_LATENCY_THRESHOLD="250" REDIS_HOST="localhost:6379" LOGSTASH_HOST="logstash.mine" LOGSTASH_PORT="8515" PROJECT="myapp" go run main.go

redis-healthy's Issues

Can this be made to work as a telegraf plugin as well?

This would be useful for users of both redis and influxdb