Closed dstathis closed 4 months ago
Just to complement the bug this unit label issue seems to be happening when a subordinate charm like hardware-observer
relates to grafana-agent
. E.g:
Model Controller Cloud/Region Version SLA Timestamp
test-charm-z6pi overlord localhost/localhost 2.9.43 unsupported 09:37:08-03:00
App Version Status Scale Charm Channel Rev Exposed Message
grafana-agent blocked 1 grafana-agent edge 52 no send-remote-write: off, grafana-cloud-config: off, logging-consumer: off
hardware-observer active 1 hardware-observer 0 no Unit is ready
ubuntu 22.04 active 1 ubuntu stable 24 no
Unit Workload Agent Machine Public address Ports Message
ubuntu/0* active idle 0 10.101.82.225
grafana-agent/0* blocked idle 10.101.82.225 send-remote-write: off, grafana-cloud-config: off, logging-consumer: off
hardware-observer/1* active idle 10.101.82.225 Unit is ready
Machine State Address Inst id Series AZ Message
0 started 10.101.82.225 juju-c1881f-0 jammy Running
On revision 28
, grafana-agent generates the config file (/etc/grafana-agent.yaml) that includes the unit label of ubuntu/0
, but this does not happen on newer releases.
With the following deployment:
╭─ubuntu@charm-dev-juju-34 ~ [lxd:hwo]
╰─$ jst
Model Controller Cloud/Region Version SLA Timestamp
hwo lxd localhost/localhost 3.4.0 unsupported 16:45:56-03:00
SAAS Status Store URL
grafana-dashboards active microk8s admin/cos.grafana-dashboards
loki-logging active microk8s admin/cos.loki-logging
prometheus-receive-remote-write active microk8s admin/cos.prometheus-receive-remote-write
App Version Status Scale Charm Channel Rev Exposed Message
agent active 1 grafana-agent edge 54 no
hwo active 1 hardware-observer edge 38 no Unit is ready
ubuntu 22.04 active 1 ubuntu stable 24 no
Unit Workload Agent Machine Public address Ports Message
ubuntu/0* active idle 0 10.10.106.208
agent/3* active idle 10.10.106.208
hwo/0* active idle 10.10.106.208 Unit is ready
Machine State Address Inst id Base AZ Message
0 started 10.10.106.208 juju-261ede-0 ubuntu@22.04 Running
Integration provider Requirer Interface Type Message
agent:grafana-dashboards-provider grafana-dashboards:grafana-dashboard grafana_dashboard regular
agent:peers agent:peers grafana_agent_replica peer
hwo:cos-agent agent:cos-agent cos_agent subordinate
loki-logging:logging agent:logging-consumer loki_push_api regular
prometheus-receive-remote-write:receive-remote-write agent:send-remote-write prometheus_remote_write regular
ubuntu:juju-info agent:juju-info juju-info subordinate
ubuntu:juju-info hwo:general-info juju-info subordinate
The metrics
section in grafana-agent is:
╭─ubuntu@charm-dev-juju-34 ~ [lxd:hwo]
╰─$ juju ssh agent/3 "cat /etc/grafana-agent.yaml"
...
metrics:
configs:
- name: agent_scraper
remote_write:
- tls_config:
insecure_skip_verify: false
url: http://192.168.1.250/cos-prometheus-0/api/v1/write
scrape_configs:
- job_name: hwo_0_default
metrics_path: /metrics
static_configs:
- labels:
juju_application: hwo
juju_model: hwo
juju_model_uuid: 71de48c1-0f16-40b7-8ed3-87748d261ede
targets:
- localhost:10200
...
After adding one more unit to the ubuntu
application:
╭─ubuntu@charm-dev-juju-34 ~ [lxd:hwo]
╰─$ jst
Model Controller Cloud/Region Version SLA Timestamp
hwo lxd localhost/localhost 3.4.0 unsupported 17:02:05-03:00
SAAS Status Store URL
grafana-dashboards active microk8s admin/cos.grafana-dashboards
loki-logging active microk8s admin/cos.loki-logging
prometheus-receive-remote-write active microk8s admin/cos.prometheus-receive-remote-write
App Version Status Scale Charm Channel Rev Exposed Message
agent active 2 grafana-agent edge 54 no
hwo active 2 hardware-observer edge 38 no Unit is ready
ubuntu active 2 ubuntu stable 24 no
Unit Workload Agent Machine Public address Ports Message
ubuntu/0* active idle 0 10.10.106.208
agent/3* active idle 10.10.106.208
hwo/0* active idle 10.10.106.208 Unit is ready
ubuntu/2 active idle 2 10.10.106.44
agent/4 active idle 10.10.106.44
hwo/3 active idle 10.10.106.44 Unit is ready
Machine State Address Inst id Base AZ Message
0 started 10.10.106.208 juju-261ede-0 ubuntu@22.04 Running
2 started 10.10.106.44 juju-261ede-2 ubuntu@22.04 Running
Integration provider Requirer Interface Type Message
agent:grafana-dashboards-provider grafana-dashboards:grafana-dashboard grafana_dashboard regular
agent:peers agent:peers grafana_agent_replica peer
hwo:cos-agent agent:cos-agent cos_agent subordinate
loki-logging:logging agent:logging-consumer loki_push_api regular
prometheus-receive-remote-write:receive-remote-write agent:send-remote-write prometheus_remote_write regular
ubuntu:juju-info agent:juju-info juju-info subordinate
ubuntu:juju-info hwo:general-info juju-info subordinate
The metrics
section in both grafana-agent config files is the same:
╭─ubuntu@charm-dev-juju-34 ~ [lxd:hwo]
╰─$ juju ssh agent/3 "cat /etc/grafana-agent.yaml"
...
metrics:
configs:
- name: agent_scraper
remote_write:
- tls_config:
insecure_skip_verify: false
url: http://192.168.1.250/cos-prometheus-0/api/v1/write
scrape_configs:
- job_name: hwo_0_default
metrics_path: /metrics
static_configs:
- labels:
juju_application: hwo
juju_model: hwo
juju_model_uuid: 71de48c1-0f16-40b7-8ed3-87748d261ede
targets:
- localhost:10200
...
╭─ubuntu@charm-dev-juju-34 ~ [lxd:hwo]
╰─$ juju ssh agent/4 "cat /etc/grafana-agent.yaml"
...
metrics:
configs:
- name: agent_scraper
remote_write:
- tls_config:
insecure_skip_verify: false
url: http://192.168.1.250/cos-prometheus-0/api/v1/write
scrape_configs:
- job_name: hwo_0_default
metrics_path: /metrics
static_configs:
- labels:
juju_application: hwo
juju_model: hwo
juju_model_uuid: 71de48c1-0f16-40b7-8ed3-87748d261ede
targets:
- localhost:10200
...
Bug Description
https://github.com/canonical/hardware-observer-operator/issues/158
To Reproduce
-
Environment
-
Relevant log output
Additional context
No response