how does healthcheckbot scales ?

[thomasleveil]

Hi, I'm wondering how this project scale. I'm looking for a tool that will need to monitor a little less than a thousand webpages every 5 minute.

Can healthcheckbot handle that ?

[corvis]

Hi @thomasleveil

There are no any specific scaling mechanisms integrated into the tool however you can easily divide your list of webpages into smaller pieces and deploy on on separate VMs. Instances of healthcheckbot are absolutely independent from each other so there is no need to configure and maintain a cluster.

You might want to have something like this:

config1.yaml

---
app:
  id: mybot-1

outputs:
  gelf:
    provider: healthcheckbot.outputs.GelfOutput
    gelf_host: graylog.lan
    gelf_port: 12201
    gelf_protocol: udp

triggers:
  each_5_minutes:
    provider: healthcheckbot.triggers.SimpleTimer
    interval: 300

watchers:
  my_page_1:
    provider: healthcheckbot.watchers.HttpRequest
    url: http://mysite1.com/page1
    assert_response_time: 2
    assert_status: 200
    triggers:
      - each_5_minutes
...................
  my_page_200:
    provider: healthcheckbot.watchers.HttpRequest
    url: http://mysite5.com/page200
    assert_response_time: 2
    assert_status: 200
    triggers:
      - each_5_minutes

and on another machine you could have very similar config with the only difference in target urls:

config2.yaml

---
app:
  id: mybot-2

outputs:
  gelf:
    provider: healthcheckbot.outputs.GelfOutput
    gelf_host: graylog.lan
    gelf_port: 12201
    gelf_protocol: udp

triggers:
  each_5_minutes:
    provider: healthcheckbot.triggers.SimpleTimer
    interval: 300

watchers:
  my_page_201:
    provider: healthcheckbot.watchers.HttpRequest
    url: http://mysite21.com/page1
    assert_response_time: 2
    assert_status: 200
    triggers:
      - each_5_minutes
...................
  my_page_400:
    provider: healthcheckbot.watchers.HttpRequest
    url: http://mysite20.com/page400
    assert_response_time: 2
    assert_status: 200
    triggers:
      - each_5_minutes

Examples above shows how to setup distributed monitoring system which reports results into single Graylog server so you have one single place where data lives.

Hope this helps.