Couldn't connect to redis instance

Question

Couldn't connect to redis instance

KYV365 opened this issue 2 years ago · 6 comments

Describe the problem
redis three master，three slave
192.168.0.1 7001 master
192.168.0.1 7002 slave
192.168.0.2 7001 master
192.168.0.2 7002 slave
192.168.0.3 7001 master
192.168.0.3 7002 slave
./redis_exporter -redis.addr 192.168.0.1 7001 -redis.password xxx -web.listen-address :9121

report error:
INFO[0000] Redis Metrics Exporter v1.40.0 build date: 2022-06-10-00:53:59 sha1: 3a74303 Go: go1.18.3 GOOS: linux GOARCH: amd64
INFO[0000] Providing metrics at :9121/metrics
ERRO[0000] Couldn't connect to redis instance (redis://192.168.0.1 7001)
ERRO[0001] Couldn't connect to redis instance (redis://192.168.0.1 7002)
ERRO[0002] Couldn't connect to redis instance (redis://192.168.0.2 7001)
ERRO[0004] Couldn't connect to redis instance (redis://192.168.0.2 7002)
ERRO[0005] Couldn't connect to redis instance (redis://192.169.6.9:7003)
ERRO[0006] Couldn't connect to redis instance (redis://192.168.0.3 7001)
ERRO[0007] Couldn't connect to redis instance (redis://192.168.0.3 7002)

What version of redis_exporter are you running?
Please run redis_exporter --version if you're not sure what version you're running.
redis_exporter-v1.40.0
Redis Version: 6.0.9

Answer 1 · 2023-03-03T09:30:20.000Z

redis_exporter-v1.40.0.linux-amd64.tar
prometheus-2.41.0.linux-amd64.tar

Answer 2 · 2023-03-06T03:40:41.000Z

redis_version:5.0.7
redis_exporter-v1.48.0.linux-amd64
I have some issues just like raojun0504
In a redis cluster built on 5 instances and each instance runs master and slave on different ports and the whole cluster has credential authentication.

172.16.1.21 7000 master
172.16.1.21 7002 slave
172.16.1.22 7000 master
172.16.1.22 7001 slave
172.16.1.23 7000 master
172.16.1.23 7001 slave
172.16.1.24 7000 master
172.16.1.24 7001 slave
172.16.1.25 7000 master
172.16.1.25 7001 slave

Create a redis_exporter.conf file and it contains master & slave IP and PORT + PASSWORD

redis_addr: "redis://<PASSWORD>@IP:7000"
redis_alias: "master"

redis_slave_addr: "redis://<PASSWORD>@IP:7001"
redis_slave_alias: "slave"

namespace: "redis"

When running the below command there would be errors appear:

redis_exporter -config-command redis_exporter.conf 
INFO[0000] Redis Metrics Exporter v1.48.0    build date: 2023-03-03-03:23:47    sha1: 30fba62cc31d80ccb8653ac33965e17c15929a24    Go: go1.20.1    GOOS: linux    GOARCH: amd64 
INFO[0000] Providing metrics at :9121/metrics           
ERRO[0001] Couldn't connect to redis instance (redis://localhost:6379) 
ERRO[0002] Couldn't connect to redis instance (redis://localhost:6379)

There is no redis service running on redis://localhost:6379 and when I pass the config file to redis_exporter it does not work probably.
Any suggestion?

Answer 3 · 2023-03-07T08:57:43.000Z

redis_version:5.0.7 redis_exporter-v1.48.0.linux-amd64 I have some issues just like raojun0504 In a redis cluster built on 5 instances and each instance runs master and slave on different ports and the whole cluster has credential authentication.
172.16.1.21 7000 master
172.16.1.21 7002 slave
172.16.1.22 7000 master
172.16.1.22 7001 slave
172.16.1.23 7000 master
172.16.1.23 7001 slave
172.16.1.24 7000 master
172.16.1.24 7001 slave
172.16.1.25 7000 master
172.16.1.25 7001 slave
Create a redis_exporter.conf file and it contains master & slave IP and PORT + PASSWORD
redis_addr: "redis://<PASSWORD>@IP:7000"
redis_alias: "master"

redis_slave_addr: "redis://<PASSWORD>@IP:7001"
redis_slave_alias: "slave"

namespace: "redis"
When running the below command there would be errors appear:
redis_exporter -config-command redis_exporter.conf 
INFO[0000] Redis Metrics Exporter v1.48.0    build date: 2023-03-03-03:23:47    sha1: 30fba62cc31d80ccb8653ac33965e17c15929a24    Go: go1.20.1    GOOS: linux    GOARCH: amd64 
INFO[0000] Providing metrics at :9121/metrics           
ERRO[0001] Couldn't connect to redis instance (redis://localhost:6379) 
ERRO[0002] Couldn't connect to redis instance (redis://localhost:6379) 
There is no redis service running on redis://localhost:6379 and when I pass the config file to redis_exporter it does not work probably. Any suggestion?

scrape_configs:

The job name is added as a label `job=<job_name>` to any timeseries scraped from this config.

job_name: 'redis_cluster'
static_configs:
- targets:
  - redis://127.0.0.1:7001
  - redis://127.0.0.1:7002
  - redis://127.0.0.1:7003
  - redis://127.0.0.1:7004
  - redis://127.0.0.1:7005
  - redis://127.0.0.1:7006
    metrics_path: /scrape
    relabel_configs:
- source_labels: [address]
  target_label: __param_target
- source_labels: [__param_target]
  target_label: instance
- target_label: address
  replacement: 127.0.0.1:9121
job_name: "redis_exporter"
metrics_path defaults to '/metrics'

scheme defaults to 'http'.
static_configs:
- targets:
  - 127.0.0.1:9121
  - 127.0.0.1:9122
  - 127.0.0.1:9123
  - 127.0.0.1:9124
  - 127.0.0.1:9125
  - 127.0.0.1:9126
    the six redis_exporter need to be started for redis_cluster，one to one。it is not seem to support redis_cluster smoothly。maybe the grafana and redis-datasource is suitable。

Answer 4 · 2023-03-07T08:58:01.000Z

redis_version:5.0.7 redis_exporter-v1.48.0.linux-amd64 I have some issues just like raojun0504 In a redis cluster built on 5 instances and each instance runs master and slave on different ports and the whole cluster has credential authentication.
172.16.1.21 7000 master
172.16.1.21 7002 slave
172.16.1.22 7000 master
172.16.1.22 7001 slave
172.16.1.23 7000 master
172.16.1.23 7001 slave
172.16.1.24 7000 master
172.16.1.24 7001 slave
172.16.1.25 7000 master
172.16.1.25 7001 slave
Create a redis_exporter.conf file and it contains master & slave IP and PORT + PASSWORD
redis_addr: "redis://<PASSWORD>@IP:7000"
redis_alias: "master"

redis_slave_addr: "redis://<PASSWORD>@IP:7001"
redis_slave_alias: "slave"

namespace: "redis"
When running the below command there would be errors appear:
redis_exporter -config-command redis_exporter.conf 
INFO[0000] Redis Metrics Exporter v1.48.0    build date: 2023-03-03-03:23:47    sha1: 30fba62cc31d80ccb8653ac33965e17c15929a24    Go: go1.20.1    GOOS: linux    GOARCH: amd64 
INFO[0000] Providing metrics at :9121/metrics           
ERRO[0001] Couldn't connect to redis instance (redis://localhost:6379) 
ERRO[0002] Couldn't connect to redis instance (redis://localhost:6379) 
There is no redis service running on redis://localhost:6379 and when I pass the config file to redis_exporter it does not work probably. Any suggestion?

scrape_configs:

The job name is added as a label `job=<job_name>` to any timeseries scraped from this config.

job_name: 'redis_cluster'
static_configs:
- targets:
  - redis://127.0.0.1:7001
  - redis://127.0.0.1:7002
  - redis://127.0.0.1:7003
  - redis://127.0.0.1:7004
  - redis://127.0.0.1:7005
  - redis://127.0.0.1:7006
    metrics_path: /scrape
    relabel_configs:
- source_labels: [address]
  target_label: __param_target
- source_labels: [__param_target]
  target_label: instance
- target_label: address
  replacement: 127.0.0.1:9121
job_name: "redis_exporter"
metrics_path defaults to '/metrics'

scheme defaults to 'http'.
static_configs:
- targets:
  - 127.0.0.1:9121
  - 127.0.0.1:9122
  - 127.0.0.1:9123
  - 127.0.0.1:9124
  - 127.0.0.1:9125
  - 127.0.0.1:9126
    the six redis_exporter need to be started for redis_cluster，one to one。it is not seem to support redis_cluster smoothly。maybe the grafana and redis-datasource is suitable。

redis_version:5.0.7 redis_exporter-v1.48.0.linux-amd64 I have some issues just like raojun0504 In a redis cluster built on 5 instances and each instance runs master and slave on different ports and the whole cluster has credential authentication.
172.16.1.21 7000 master
172.16.1.21 7002 slave
172.16.1.22 7000 master
172.16.1.22 7001 slave
172.16.1.23 7000 master
172.16.1.23 7001 slave
172.16.1.24 7000 master
172.16.1.24 7001 slave
172.16.1.25 7000 master
172.16.1.25 7001 slave
Create a redis_exporter.conf file and it contains master & slave IP and PORT + PASSWORD
redis_addr: "redis://<PASSWORD>@IP:7000"
redis_alias: "master"

redis_slave_addr: "redis://<PASSWORD>@IP:7001"
redis_slave_alias: "slave"

namespace: "redis"
When running the below command there would be errors appear:
redis_exporter -config-command redis_exporter.conf 
INFO[0000] Redis Metrics Exporter v1.48.0    build date: 2023-03-03-03:23:47    sha1: 30fba62cc31d80ccb8653ac33965e17c15929a24    Go: go1.20.1    GOOS: linux    GOARCH: amd64 
INFO[0000] Providing metrics at :9121/metrics           
ERRO[0001] Couldn't connect to redis instance (redis://localhost:6379) 
ERRO[0002] Couldn't connect to redis instance (redis://localhost:6379) 
There is no redis service running on redis://localhost:6379 and when I pass the config file to redis_exporter it does not work probably. Any suggestion?

scrape_configs:

The job name is added as a label `job=<job_name>` to any timeseries scraped from this config.

job_name: 'redis_cluster'
static_configs:
- targets:
  - redis://127.0.0.1:7001
  - redis://127.0.0.1:7002
  - redis://127.0.0.1:7003
  - redis://127.0.0.1:7004
  - redis://127.0.0.1:7005
  - redis://127.0.0.1:7006
    metrics_path: /scrape
    relabel_configs:
- source_labels: [address]
  target_label: __param_target
- source_labels: [__param_target]
  target_label: instance
- target_label: address
  replacement: 127.0.0.1:9121
job_name: "redis_exporter"
metrics_path defaults to '/metrics'

scheme defaults to 'http'.
static_configs:
- targets:
  - 127.0.0.1:9121
  - 127.0.0.1:9122
  - 127.0.0.1:9123
  - 127.0.0.1:9124
  - 127.0.0.1:9125
  - 127.0.0.1:9126
    the six redis_exporter need to be started for redis_cluster，one to one。it is not seem to support redis_cluster smoothly。maybe the grafana and redis-datasource is suitable。

Answer 5 · 2023-03-17T10:38:50.000Z

I had the same issue and as far as I get it right, this error messages comes from the redis exporter itself.

I've adapted my docker compose file with the following line.

    command: ["--redis.addr=<string: one_of_my_host_or_ip>:<int: the_port>"]  # This line prevents the logging entry `time="2023-03-17T10:09:20Z" level=error msg="Couldn't connect to redis instance (redis://localhost:6379)"` ref: https://github.com/oliver006/redis_exporter#prometheus-configuration-to-scrape-multiple-redis-hosts

It is a bit odd, that you have to choose one redis server that should never go down to prevent this error log message. If there is a better solution, let me know.

Cheers,
Stev

Edit

@raojun0504 feel free to close this issue if my answer solves your problem (neither you like it or not ;-))

Answer 6 · 2023-03-23T02:51:05.000Z

It's in the README: https://github.com/oliver006/redis_exporter#prometheus-configuration-to-scrape-multiple-redis-hosts

Run the exporter with the command line flag --redis.addr= so it won't try to access the local instance every time the /metrics endpoint is scraped

The job name is added as a label job=<job_name> to any timeseries scraped from this config.

metrics_path defaults to '/metrics'

scheme defaults to 'http'.

The job name is added as a label job=<job_name> to any timeseries scraped from this config.

metrics_path defaults to '/metrics'

scheme defaults to 'http'.

The job name is added as a label job=<job_name> to any timeseries scraped from this config.

metrics_path defaults to '/metrics'

scheme defaults to 'http'.

Edit

The job name is added as a label `job=<job_name>` to any timeseries scraped from this config.

The job name is added as a label `job=<job_name>` to any timeseries scraped from this config.

The job name is added as a label `job=<job_name>` to any timeseries scraped from this config.