pinGOr

Create logs about disconnections with Golang. Run pinGOr and see it's logs to know if your internet connection was interrupted or not. Currently pinGOr can log all information to console, persist reconnection events to DynamoDB and open /metrics endpoint for Prometheus

How?

PinGOr will read provided config and try to:

• resolve provided host names to IPs,
• inspect HTTP statuses for provided URLs.

It will start checking the connection after configured minimal checking period and if the connection will be ok the period will be doubled. When connection checks will drop below configured success rate threshold, then the connection will be marked as "dropped" and proper log will be created.

Usage

In order to build the executable run:

GO111MODULE=on go build

In order to run the executable:

./pingor -debug -config config.yaml

In order to test it:

go test ./... -tags unit

Config

success_rate_threshold: 0.74  # rate of successfull sub-checks to mark whole check as successfull
success_time_threshold: 5s    # Max average time of sub-checks to mark whole check as successfull
single_check_timeout: 10s     # Timeout for single sub-check
minimal_checking_period: 1m   # Minimal, starting period for periodic checks. Will double after success
maximal_checking_period: 30m  # Maximal period for periodic checks
persister:
  dynamodb:
    enabled: false            # Store reconnection events in DynamoDB?
    region: eu-west-1         # Which AWS region should be used?
    table_name: pingor        # Which table?
    device_name: pc1          # How to call this device in DynamoDB?
http:
  urls:     # URLs to check if HTTP status is 200 OK
    - https://wp.pl
    - https://www.onet.pl
    - https://www.google.com
    - https://www.upc.pl
dns:
  hosts:  # Hosts to resolve in order to confirm connection to DNS is working. Leave empty to skip DNS checks
    - wp.pl
    - onet.pl
    - google.com
    - upc.pl
    - mbank.pl

metrics:
  enabled: false    # Serve /metrics endpoint for Prometheus?
  port: 9111

Recommended usage

Recommended usage is by adding pinGOr to systemd. Below you can see example pingor.service file, that can be helpful

[Unit]
Description=pinGOr: logging internet connectivity
After=network.target

[Service]
Type=simple
User=somebody # EDIT THIS
WorkingDirectory=/full/path/to/pingor/dir # EDIT THIS
ExecStart=/full/path/to/pingor/exec # EDIT THIS
Restart=on-abort
Environment=AWS_ACCESS_KEY_ID=....YOUR KEY..... # EDIT THIS
Environment=AWS_SECRET_ACCESS_KEY=....YOUR SECRET..... # EDIT THIS

[Install]
WantedBy=multi-user.target

After configuring it you can inspect the logs (or AWS DynamoDB) to check for connection disruption or connect /metrics endpoint with Prometheus by adding this to it's config:

scrape_configs:
  # some other jobs

  - job_name: 'pingor'
    static_configs:
      - targets: ['127.0.0.1:9111'] # Port needs to be the same as configured in pingor

Prometheus metrics

PingGOr metrics endpoint is exposing golang related metrics and network stability metrics:

• http_pingor_success_rate - 0-100 percentage rate of success
• http_pingor_response_times_seconds - seconds it took for each host to check for HTTP stauts
• dns_pingor_response_times_seconds - seconds it took for each host to resolve to IP

Using AWS DynamoDB

In order to persist reconnection events to AWS DynamoDB you have to specify your access and secret keys as environment variables. If you are running pinGOr with the help of systemd you can specify them in pingor.service file.

Why?

I have signed SLA agreement with my ISP, but didn't have any tool to actually know if there was any connection-related problem, while the PC was running and I was away.

Known issues

• Was not tested on Windows

Contributing

All ideas and pull requests are welcomed and appreciated. If you have any problem with usage don't hesitate to create an issue, we can figure out your problem together.

Result in Grafa

You could configure Grafana to communicate with Prometheus and display the metrics in beautiful way

Author

Krzysztof Gzocha