Pinger - Automated 'ping' to a server & email on failure

Image repository Dockerhub: mlefkon/pinger , Source repository Github: mlefkon/pinger

What it does

A docker container cron job runs curl to 'ping' your server/endpoint. An email is sent via SMTP relay if any problems occur.

Configuration (Environment Variables)

• Target Server
  • ENDPOINT_NAME: default: "Pinger", used for email 'subject' & logfile names about this Pinger instance
  • PING_URL: required, any valid curl URL. Include 'https://' if needed as curl defaults to http.
  • ALLOW_INSECURE: default: 0, allow for invalid certificate (set to '1')
  • IPv6: default: 0, url is in IPv6 format (set to '1')
  • EXPECTED_RESPONSE: required, expected to be found within text body returned from PING_URL
  • INTERVAL_MIN: default: 5, minutes between pings
  • RELIABLE_REFERENCE_PING_HOST: required, for connectivity test in case of PING_URL failure. Must be a ping-responsive host, not a URL. ping is used, not curl, so only domain name is needed here.
• Mail Relay
  • RELAY_HOST: required, format (incl sq brackets): [relay.host.tld]:port
  • RELAY_USERNAME: required, user's login to relay mail host
  • RELAY_PASSWORD: required
  • RELAY_SENDER_EMAIL_ADDRESS: required, should be permitted to send mails by relay host, often same as username
  • RELAY_SENDER_INFORMAL_NAME: default: Pinger
• Notification
  • TO_EMAIL_ADDR: required, recipient of alert emails
  • THRESHOLD_FAILS_FOR_EMAIL: default: 1, num of failures before an alert email is sent
  • STATUS_EMAIL_DAYS default: 30, Status report email will be send every X days

Status Report

At the end of each ${STATUS_EMAIL_DAYS} time period and email will be sent with:

• Start/End Dates
• Total down time (& times of failures)
• Percent up
• Median/Average ping time

Historical Data

Data is saved to ${ENDPOINT_NAME}.{ping|fails|history}.log files in /var/log/pinger/. A docker volume can be mounted here to persist. These are CSV files with the formats:

• Pings (~.ping.curr.log, has only current period. This is moved to ~.ping.prior.log after status report is sent.)
  • Unix Timestamp
  • Result (1- endpoint is up, 0- endpoint is down)
  • Ping Response Time
• Summary History (~.summary.history.log)
  • From Date
  • To Date
  • Up Time Percentage
  • Median Ping Time
  • Average Ping Time
• Fails (~.num.fails.log)
  • Number of current fails (reset to zero after a success)

Run

    $ docker run --name mypinger -d \
    -e ENDPOINT_NAME="My Pinger" \
    -e INTERVAL_MIN=3 \
    -e THRESHOLD_FAILS_FOR_EMAIL=2 \
    -e PING_URL=https://my.url.com/myscript \
    -e ALLOW_INSECURE=0 \
    -e IPv6=0 \
    -e RELIABLE_REFERENCE_PING_HOST=google.com \
    -e EXPECTED_RESPONSE="eg: Ping succeeded" \
    -e RELAY_HOST=smtps://smtp.relay.com \
    -e RELAY_USERNAME=user \
    -e RELAY_PASSWORD=pass \
    -e RELAY_SENDER_EMAIL_ADDRESS=useremail@relay.com \
    -e RELAY_SENDER_INFORMAL_NAME="Pinger Alert" \
    -e TO_EMAIL_ADDR=recipient@email.com \
    -e STATUS_EMAIL_DAYS=7 \
    mlefkon/pinger 

IPv6

The container should probably be run on the 'host' network, assuming the host can operate with IPv6 that the 'ping' is to outside the network. Otherwise a docker 'bridge' network will need to be used with IPv6 enabled, adding an extra layer of complexity.

Note that IPv6 addresses used with curl should be enclosed in [square brackets] like:

https://[::1]