Jump to content

Nagios like plugin infrastructure


Brian Lakstins
 Share

Recommended Posts

I monitor my infrastructure using Nagios running on multiple Linux servers.  I'm switching over to using Pulseway.

There are a lot of things that Nagios does that Pulseway does not, so I'm looking to implement some of the important things using plugins.

For starters, I'm planning on checking HTTP (speed of response, size of response, does the response contain key information).  I want to expand to other network service checks (DNS, SMTP, IMAP) to ensure my infrastructure is always available.

I am going to have multiple host and service checks run through the PluginDataCheck method of the ClientPlugin.

I'm concerned that I might need to make these checks run concurrently so that I don't block the PluginDataCheck method and make it take a long time.  They are network service checks, so if there are problems some could take up to their timeout (like 30 seconds).  Is this a valid concern?  Is there a certain amount of time that PluginDataCheck needs to run in before it times out?

My next concern was organization and notifications. If multiple checks fail, I'd like to get multiple notifications.  This is currently limited to 1 per plugin.

I was thinking that I could use the Cloud API to create a service for every check that I run.  That way each one could send it's own notifications and I could also organize the checks into groups instead of having them related to the system that's doing the monitoring.  Would that be an abuse of the Cloud API?

 

Link to comment
Share on other sites

  • Administrators

Hi Brian,

Welcome to the Pulseway community and Merry Christmas :lol:.

The 'PluginDataCheck' method will be invoked from a background thread so it will not influence other Pulseway internal monitors and it will never timeout. Plugins are not limited to 1 active notification per instance, there is an overload to the SendNotificationToAllDevices method which accepts a boolean flag that controls whether the notification can recur or not.

In regards to the Cloud/REST APIs, you wouldn't be 'abusing' the system. You are free to use all exposed methods in our APIs to your own liking, we enforce our rate limits on the server side but they are really high and configured to only trigger when something is definitely wrong in an implementation (I.e. infinite loop of repeating notifications).

-Paul

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

×
×
  • Create New...