Translations of this page:

Osmius

What is distributed monitoring?

Monitoring something means measure this thing in order to detect problems. With distributed monitoring the measures are taken from several places because the result could change.

Osmius is able to do this in a very easy way, because you can deploy a distributed Osmius infrastructure, installing Osmius master agents on several machines that sends information to Osmius central server. If Osmius master agents are installed on geographically distributed machines you will have, in fact, distributed monitoring.

Distributed monitoring example

Suppose this: we want to monitor a web page hosted at any server. Using distributed monitoring we can check availability, download speed and load time of web page from several places in on the planet (you need several Osmius master agents installed on machines located in several places on the planet). This way you understand easily where are the problems; if this web page is an online book shop you know quicly what places, countries or regions are not able to access the shop from. The business view is very interesting because you can translate it directly to lost profits.

In this case, we could monitor this web page from China, UUEE and Europe, as shown in the graph.

Distributed monitoring using Osmius

How can I do this using Osmius?

It is very easy using Osmius. Follow these steps:

  1. Install Osmius master agents on geographically (or network) distributed machines. Example: On a server located at China, UUEE and/or Europe.

    Installation Chapter
  2. Add an instance for each Osmius master agent. Example: WebChina (to monitor the web from China), WebUUEE (to monitor from United States), WebEuro (from Europe), … as many as you need.
    Adding instances
  3. Deploy agent for each instance type to those Osmius master agents. Remember, to deploy a master agent go to: Infrastructure - right click over the master agent - Deploy
    Deploying agents
  4. Match each instance with its master agent
    Matching instances and master agents
  5. Group the instances in the same service to a better organization. You can better determine the root cause of problems.
    Grouping instances into a service
  6. You can set propagation rules for a fine tunning.
  7. Don't forget set the instances locations in Maps tab
 
en/howto/002monitorizacion_distribuida.txt · Last modified: 2010/12/22 13:22 (external edit)
 
Recent changes RSS feed Creative Commons License Donate Powered by PHP Valid XHTML 1.0 Valid CSS Driven by DokuWiki