Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!


Free/Open source uptime monitoring tool that supports multiple nodes?
New on LowEndTalk? Please Register and read our Community Rules.

All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.

Free/Open source uptime monitoring tool that supports multiple nodes?

jimaekjimaek Member
edited June 2014 in General

Are there any uptime monitoring tools(self-hosted) that can do testing from multiple locations?
Basically 1 server that aggregates all the data and an unlimited amount of nodes around the world that do testing of URLs configured on the server.
If one of the nodes fails the server registers the downtime even if nodes in other locations report that everything is ok.

Basically im trying to monitor a CDN.

Comments

  • Nagios

    Thanked by 1lukesUbuntu
  • jimaekjimaek Member

    Nagios sounds like overkill for something simple as this.

  • meatonmeaton Member

    http://www.statuscake.com has a free tier that allows unlimited hosts to be monitored. The check rate is 5 minutes and is done from a random location. The paid subscription obviously will allow for checks from multiple locations, and the check rate is reduced (1 minute, 30 second/constant).

  • BradBrad Member

    I believe he wants a self hosted solution.

  • BrianHarrisonBrianHarrison Member, Patron Provider

    @meaton said:
    http://www.statuscake.com has a free tier that allows unlimited hosts to be monitored. The check rate is 5 minutes and is done from a random location. The paid subscription obviously will allow for checks from multiple locations, and the check rate is reduced (1 minute, 30 second/constant).

    OP is looking for a self-hosted solution. I may be mistaken, but it appears that StatusCake is only provided as a service.

  • blackblack Member
    edited June 2014

    I wrote a ping monitor in bash awhile back. Nothing fancy but it works. check my github in my signature :)

    Of course you can change the check condition to something else like curl -I someURL | grep 200 -v

  • perennateperennate Member, Host Rep
    edited June 2014

    I wrote this one: https://github.com/uakfdotb/bearmon

    there's nearly no documentation (the README is wrong about the SQL structure, | isn't delimiter anymore) but it is designed to be very simple and does support multiple locations.

    edit: actually it sounds like you really want single-node monitoring, but installed on several nodes; you could use @black solution and modify it to alert the master via HTTP or something?

  • jimaekjimaek Member

    @perennate Not sure what is the difference. I want 1 master server where all data gets combined and I can add servers to monitor. And an unlimited amount of nodes that will do the actual testing and then send the data to the master server.
    It can be a simple bash script or php or anything really as long as it does what I want.

    Any ideas? None of the solutions here seem to be what I need.

  • edited June 2014

    Have you tried xymon (aka hobbit)?

  • wychwych Member

    NodeQuery?

  • @wych said:
    NodeQuery?

    The OP is looking for a self hosted solution.

  • I too wrote one. However I find it tedious to maintain documentation on how to set it up. Check my sig.

  • wychwych Member

    Fair point, missed that.

  • edited June 2014

    @joelgm said:
    I too wrote one. However I find it tedious to maintain documentation on how to set it up. Check my sig.

    Nice, I like it!

    The ports being displayed for the world to see scares me, but good Job!

  • I have been looking for something that is self hosted like this for prob atleast 2 years. The closest thing that would work for you is smokeping http://oss.oetiker.ch/smokeping/

  • DroidzoneDroidzone Member
    edited June 2014

    @globalRegisters said:
    The ports being displayed for the world to see scares me, but good Job!

    Nobody but you knows what IPs are behind each alias. How many minutes does it take to do a port scan against an ip with nmap?

    Anyway, if you wish to keep it private you just need to import one of the php authentication plugins.

  • perennateperennate Member, Host Rep

    jimaek said: Not sure what is the difference. I want 1 master server where all data gets combined and I can add servers to monitor. And an unlimited amount of nodes that will do the actual testing and then send the data to the master server. It can be a simple bash script or php or anything really as long as it does what I want.

    Most of the monitoring tools that support multiple nodes, use the multiple nodes to load balance the monitoring and also to continue monitoring in case one of the nodes fails.

  • NekkiNekki Veteran

    @meaton said:
    stuff

    Are you by any chance a provider who will be offering VPS plans sometime in the next 7-10 days?

  • agentmishraagentmishra Member, Host Rep

    why dont people use the @bluevm uptime script, you could offcourse do a bit of tweak but thats the nicest one, i feel...

  • cassacassa Member
    edited June 2014

    @agentmishra said:
    why dont people use the bluevm uptime script, you could offcourse do a bit of tweak but thats the nicest one, i feel...

    Because:
    - You need to keep the page open
    - No alerts
    - Webserver on every server

  • agentmishraagentmishra Member, Host Rep

    @cassa said:
    Because: - You need to keep the page open - No alerts - Webserver on every server

    so i said, i few tweaks, a few only...

  • wojonswojons Member

    i suggested the same one but i guess no one likes it

  • meatonmeaton Member

    Nekki said: Are you by any chance a provider who will be offering VPS plans sometime in the next 7-10 days?

    I'm already a provider who offers VPS plans today. :)

  • jimaekjimaek Member

    Started a project for my exact needs https://github.com/jsdelivr/monitoring
    If anyone wants to contribute feel free to. I suck as a dev :)

    The idea is to build a stable node.php and then start deploying it around the globe.
    The data is pushed to keen.io.
    Then our load balancing algorithm will pull the information from keen.io and use it as live data to ensure the availability of the providers.
    This data is going to be on top of the existing RUM availability tests and integrated monitoring from a single location.

    Once its done we are going to h ave 3 different data points for an even more stable CDN.

Sign In or Register to comment.