Nagios and Zabbix vs SATELLIZ monitoring solution: what are the differences?

nagios zabbix satellizAs recent exhibitors at the IT Partners 2013 Show, we have had the opportunity to meet many proximity IT service providers and discuss their own monitoring projects or those they manage for their customers.

One of the most frequently-discussed topics is the difference between the SATELLIZ monitoring software solution and the more conventional open-source (and free) monitoring software such as the widely-used Nagios and Zabbix.

The final customer’s stated need is almost always the same each time.

A typical request…

“We have our own servers [on our premises], physical servers hosted by our service provider, and increasingly we have virtual servers at Amazon.” (This public cloud player appears very frequently in conversations.).

“A few years ago, we installed Nagios on our servers and it works very well, even though it was a little long and complicated to configure. Nagios is free, that is its main advantage, although it requires considerable expertise to set up, expertise that we had at the time. It is more demanding now: we must work on Nagios regularly, whenever we change our servers’ scope, but our Linux engineers are busy on other things…”

“The critical servers on our hoster’s premises are partly monitored by our hosting service provider, which provides full or partial administration of our business applications. We do not have a monitoring tool for the virtual servers provided by our public cloud service provider, though. We are again considering Nagios, … or Zabbix which seems more advanced, …”

“I have heard of Satelliz, … Can you can tell me more?”

We have exaggerated, but only slightly. Of course, each need is specific and unique to the company concerned, but this basis is very close to reality.

In fact, we find the same problems that confronted us a few years ago, when we directed a hosting company: our monitoring tools of the time could no longer productively support us in repsonse to our infrastructure changes.

Although companies have great respect for the free solutions such as Nagios or Zabbix, they have now found their limits and are open to an alternative.

  

“Free is expensive”

This is how we could summarise our discussions with our contacts.

Although these solutions are initially free, they are actually very costly, demanding much of IT experts’ time: companies can no longer afford the luxury of using their rare skills to manage an IT administration tool when, at the same time, the companies’ “business” projects (functional roadmap, deployment of new architectures, etc.) are the “business” priority and need the greatest expertise!

As a result, when companies are confronted with monitoring schedule and budget slippage due to free solutions demanding now-rare technical expertise, a solution that is simple to set up, offers controlled costs and is easy to use represents an attractive alternative.

This is a brief summary of the differences between the approaches with an in-house Nagios or Zabbix solution and the SATELLIZ server monitoring solution.

 

Nagios and Zabbix

SATELLIZ 

THE MAIN DIFFERENCES

 

Who is the solution intended for?
Who are the users?

  • Production engineers
  • Linux experts
  • Experienced technicians
  • Management
  • IT managers
  • Developers
  • Technicians, engineers and experts
 

Budget differences

Initial investment

  • Hardware cost (monitoring servers)
  • Deployment cost (man-days)

Initial investment

  • None, no installation cost
 

External recurrent costs

  • Maintenance of physical servers
  • Maintenance of installed software
  • Any hosting

Internal recurrent costs (hidden)

  • Experts to operate and maintain the solution’s configuration
  • Training and skills transfer
  • Documentation

 

 

External recurrent costs

  • Fixed monthly cost per server

Internal recurrent costs (hidden)

  • Simple solution for use by non-experts; very often used directly by developers. 
    Low costs

Considerable initial investments and hidden recurrent costs

0% Capex (investments), 100% Opex (operating costs)

Cost control and reduction:
see all the advantages 

 

Solution implementation

  • Complete project (see architecture below)
  • Long lead times (and large workload)
  • Immediate start-up
 

Technical architecture

Assumptions: 50 servers to monitor, 10 of which are within the company, 10 are hosted by a hosting service provider and 30 are virtual servers provided by a public cloud service provider. Monitoring of critical IT resources, involving a high-availability monitoring solution.
  

Central monitoring platform:

In the case of Nagios:

  • Install a Nagios server per site:
    • 1 server within the company
    • 1 server on the hosting service provider’s premises
    • 1 server on the public cloud service provider’s premises.
  • For a consolidated view of the three servers, the installation of a consolidation server (such as Centreon) is necessary.
  • Configure the 4 servers (servers, networks and security).

In the case of Zabbix:

  • Install a central Zabbix server, which will provide the consolidated view of the proxies as well as monitoring the local resources on this site, and install a Zabbix proxy per
    additional site:
    • 1 Zabbix server within the company
    • 1 Zabbix server on the hosting service provider’s premises
    • 1 Zabbix server on the public cloud service provider’s premises.
    • Configure the servers (servers, networks and security).

 

Central monitoring platform:

  • No environment to install.
  • Consolidated view via a web interface (monitoring portal).

Components on the servers to be monitored:

In the case of Nagios:

  • Deploy the Nagios tools (binaries and scripts) on each server.
  • “Customize” the packages according to the OS (OS type, version, etc.).
  • For each new sensor, deploy new components on the servers (binaries and scripts).

In the case of Zabbix:

  • Deploy the Zabbix agent on each server; the agent is configured from the Zabbix server.
  • For each new sensor, deploy new components on the servers (binaries and scripts).

 

Components on the servers to be monitored:

  • Deploy the pre-packaged agents on each server (multi-OS compatibility).
  • 30-second installation; bulk installation possible.

Agent sensor and plugin management via the web interface.  

Network and security prerequisites:

In the case of Nagios:

  • Opening communication ports (in both directions), introducing security constraints (to ensure that the monitoring servers do not become security black holes).

In the case of Zabbix:

Opening specific communication ports between the sites, introducing security constraints (many ports must be opened if there is a single Zabbix server for several sites; few must opened if there is a single server or proxy per site).

Network and security prerequisites:

  • None (the agents connect and then communicate via an outgoing HTTPS connection to the central platform, via a proxy when appropriate — the central platform does not initiate a connection to the agents).
 

Solution operation

In the case of Nagios:

  • The servers, systems, monitoring engines and databases must be managed.
  • All hardware or system changes within the scope to be monitored must be updated in the monitoring configuration.
  • The configuration must be updated, in “command-line” mode, both on the monitoring side and on the monitored server side (Nagios tools update, including in bulk, and sensor and plugin configuration update, including via templates).

In the case of Zabbix:

  • The servers, systems, monitoring engines and databases must be managed.
  • All hardware or system changes within the scope to be monitored must be updated in the monitoring configuration.
  • The configuration is automatically discovered and updated via the Zabbix web interface.

To add a specific new sensor, the configuration must be updated both on the monitoring side and on the monitored server side.

 

  • The solution is fully operated and maintained by SATELLIZ.
  • The SATELLIZ solution discovers hardware and system changes and automatically updates the sensor lists.
  • Configuration updates are made directly from the web interface (agent updates, including in bulk, and sensor and plugin configuration updates, including via templates).
 

Functionalities & Support

Virtually identical functionalities

Strengths

  • Very wide range of metric points (sensors) through the plugins made available by the community.
  • Endless monitoring “customization” possibilities (via scripting).

Strengths

  • Compatibility with all Nagios plugins.
  • Endless monitoring “customization” possibilities (Custom Plugins).
  • Advanced, multi-channel notification system.
  • Precise metrics: real- or delayed-time HD graphs with no smoothing (the peaks and the high-definition data are retained for detailed later analysis).
  • Graphical interface provides simple, mouse-click operation (no command lines).
  • Cloud account management (public/private).
  • Integrated Ticketing module.

Support

  • Documentation is highly technical, numerous and free.
  • No official support (open-source product).
  • “Community” support via forums, with no commitment.
  • Paying “Enterprise” support proposed by the editor or third-party service providers.

Support

  • 24/7 solution support.
  • Functional roadmap close to customer expectations.
  • Multi-tenant SaaS solution is always up-to-date.
  • Possibility of delegating level 1 standby (24/7 remote incident resolution).