Enable Autodiscover clients in Prometheus formula in Uyuni 2020.03

[bpetru]

Hi,
I try to configure monitoring with Prometheus, but after I activate Autodiscover clients in salt formula the Prometheus server dose not start.

The error is:
--config.file=\"/etc/prometheus/prometheus.yml\"): parsing YAML file /etc/prometheus/prometheus.yml: yaml: unmarshal errors:\n line 36: field uyuni_sd_configs not found in type config.plain" (--config.file=\"/etc/prometheus/prometheus.yml\"): parsing YAML file /etc/prometheus/prometheus.yml: yaml: unmarshal errors:\n line 36: field uyuni_sd_configs not found in type config.plain"

If I disable Autodiscover the server works ok, and I have Uyuni performance data.

[paususe]

@cavalheiro @renner
Could you please take a look?

[srbarrios]

Might it just be an indentation issue?
This is the example from the doc, but it might be wrong?

# Managed systems metrics:
- job_name: 'mgr-clients'
  uyuni_sd_configs:
   - host: "http://server.url"
     username: "admin"
     password: "admin"

[calancha]

Using the following configuration (that is, I keep the wrong indentation), it works:

# Sample config for Prometheus.

global:
  scrape_interval:     15s # By default, scrape targets every 15 seconds.
  evaluation_interval: 15s # By default, scrape targets every 15 seconds.
  # scrape_timeout is set to the global default (10s).

  # Attach these labels to any time series or alerts when communicating with
  # external systems (federation, remote storage, Alertmanager).
  external_labels:
      monitor: 'example'

# Load and evaluate rules in this file every 'evaluation_interval' seconds.
rule_files:
  # - "first.rules"
  # - "second.rules"

# {productname} self-health metrics
scrape_configs:
- job_name: 'mgr-server'
  static_configs:
    - targets:
      - 'vm153088.qa.prv.suse.net:9100'  # Node exporter
      - 'vm153088.qa.prv.suse.net:9187'  # PostgreSQL exporter
      - 'vm153088.qa.prv.suse.net:5556'  # JMX exporter (Tomcat)
      - 'vm153088.qa.prv.suse.net:5557'  # JMX exporter (Taskomatic)
      - 'vm153088.qa.prv.suse.net:9800'  # Taskomatic
    - targets:
      - 'vm153088.qa.prv.suse.net:80'    # Message queue
      labels:
        __metrics_path__: /rhn/metrics

# Managed systems metrics:
- job_name: 'mgr-clients'
  uyuni_sd_configs:
   - host: "http://vm153088.qa.prv.suse.net"
     username: "admin"
     password: "admin"

Here, the SUMA server is the one for testing beta 2, and the monitoring client is a sles15sp2.

[calancha]

Screenshot of restart service after apply config. above:

[calancha]

Prometheus endpoint after restart service w/ above config.:

[cavalheiro]

@mateiw can you take a look at this please?

[mateiw]

I couldn't reproduce the issue. It works fine for me.
@bpetru could you please post a screenshot of the Prometheus formula and the content of the /etc/prometheus/prometheus.yml file ?

[willianops]

Man, I have the same problem.. but indetation don't work for me..

error:

promtool check config prometheus.yml

Checking prometheus.yml
FAILED: parsing YAML file prometheus.yml: yaml: unmarshal errors:
line 44: field uyuni_sd_configs not found in type config.plain

Config:

global:
scrape_interval: 15s
evaluation_interval: 15s

Alertmanager configuration

alerting:
alertmanagers:

• static_configs:
  • targets:
    • localhost:9093

rule_files:

• /etc/prometheus/rules/general-rules.yml
• /etc/prometheus/rules/prometheus-rules.yml

Scrape configurations

scrape_configs:

--------------------

Monitor 001.xxx.net

--------------------

• job_name: 'mgr-server'
  static_configs:
  • targets:
    • 001.xxx.net:9100 # Node exporter
    • 001.xxx.net:5556 # Tomcat JMX
    • 001.xxx.net:5557 # Taskomatic JMX
    • 001.xxx.net:9800 # Uyuni server exporter
      labels: {}
  • targets:
    • 001.xxx.net:80 # Message queue
      labels:
      metrics_path: /rhn/metrics
  • targets:
    • 001.xxx.net:9187 # PostgresSQL
      labels:
      role: postgres

--------------------

Auto discover clients of 001.xxx.net

--------------------

• job_name: 'mgr-clients'
  uyuni_sd_configs:
  • host: https://001.xx.net
    username: admin
    password: xxxx

--------------------

Metamonitoring

--------------------

• job_name: prometheus
  static_configs:
  • targets:
    • 008.net:9090

[cavalheiro]

@willianops which Prometheus package are you using?
For service discovery to work correctly, you have to use the package provided in the Uyuni client tools channel.

[willianops]

Uyuni

I'm use Suse Manager to deploy configurations of prometheus. All configs are native and default . Of Course Suse Manager using Uyuni in background. by the way, I opened a Case in Suse Support

[noelmcloughlin]

you have to use the package provided in the Uyuni client tools channel.

Hi, I help maintain prometheus-formula and this issue is logged there too.
saltstack-formulas/prometheus-formula#47
What is the Uyuni client tools channel and should promethtus-formula support that source? (since we aim to be comprehensive).

[aaannz]

@witekest If you will have time, can you take a look at this if it is still valid and possibly close the issue?

[witekest]

you have to use the package provided in the Uyuni client tools channel.

Hi, I help maintain prometheus-formula and this issue is logged there too. saltstack-formulas/prometheus-formula#47 What is the Uyuni client tools channel and should promethtus-formula support that source? (since we aim to be comprehensive).

Uyuni service discovery is still not available in upstream Prometheus. Uyuni client tools is the repository providing the forked Prometheus with Uyuni SD. After prometheus/prometheus#8190 gets accepted, the problem should be solved.

[aaannz]

Upstream PR merged, this issue should be solved now. Closing issue.