IN DEVELOPMENT

Mongraph

Monitor infrastructure nodes and visualize it as graph

How to start

1. setup config/config.yml
2. (optional) setup static/graphs/root.mermaid and other mermaid-files based on links from config-file
3. start the server by docker compose up
4. open in browser by http://localhost:8000/

Supported chart types

• flowchart (https://mermaid.ai/open-source/syntax/flowchart.html)
• block (https://mermaid.ai/open-source/syntax/block.html)

Config sturucture:

• label: name of project
• ui_update_interval: UI update frequency in seconds
• graph_file: name of mermaid graph file in config/graph/ folder
• database_uri: local redis-compatible database address
• defaults: default values of node settings
  • update_interval: update node metric frequency in seconds
  • levels: metric value levels setting
    • normal: condition for normal-status. If it defined, then metric status mode is normal/danger.
    • warning: condition for normal-status
    • danger: condition for normal-status
    • measurement: level measurement for output
• notifications: notification settings
  • delay: time shift before to notify
  • min_status: minimum metric statis for notification
  • recipients: list of recipients. Supported: shell-command and emails (via smtp-link)
• providers: setting for metric providers
  • [name of provider]:
    • type: type of provider. Supported: prometeus and plaintext
    • metric_source: URI or group of metric_source-settings
      • only_for_inheritance: true/false - to use metric_source or leave it's as base settings for metrics
      • data_source: URI to metric data source or group of it parts
        • uri: any supported link. Look at sample of URI syntax group
        • query: query-part of uri (if it is http/https type). Primary for inheritance cases (useful in metrics part).
        • body: body-part or uri (if it is http/https type)
      • request_timeout: request timeout in seconds
      • update_interval: reload data frequency in seconds
      • mask_re: perl regular expression with possible to extract: datetime, name, value, location and details. Example: "^(?P<datetime>[^ ] [^ ])\s+(?P<name>[^\s]+)\s+(?P<value>[^\s]+)"
      • node_name_attr: name of data from mask_re which will be used as node name (group metrics by nodes)
      • value_source: specific source of metric value. Supported: http-code, exit-code
      • datetime_format: datetime pattern for datetime of data from mask_re
      • rows_filter: condition
    • metrics: list of metric settings
      • [name of metric]:
        • metric_source: url or group of metric_source-settings (with inheritance of provider's block. filter downloaded data by provider or make request by modified url)
        • levels: metric value levels setting for provider metric (structure same as in defaults)
• nodes: setting for nodes tree
  • [node name]: node name
    • label: label of node
    • graph_file: name of mermaid graph file of node childs in config/graph/ folder
    • child_nodes_from_provider: list of links to provider metric
      • [provider://iac-prometeus?filter=(location!='private')&node_name_attr=name,instance]
    • child_nodes: list of child nodes
      • [child node name]: child node name
        • metric_source: URI or group of metric_source-settings (structure same as for providers)
        • levels: metric value levels setting for provider metric (structure same as in defaults)

sample of URI syntax:

• https://google.com/
• file:///mnt/host_syslog
• shell://nc -z valkey 6379
• smtp://login:pasword@stmp.server:port?from=noreply@foo.com&to=user@foo.com