Nagios Blessing or Curse

[RobertFontaine]

I am in the midst of installing and configuring Nagios in my dungeon lab ( A mix of virtualized servers, assorted hardware, router, switch and growing ). I was looking at Zabbix and thinking that it looked kind of nice but then I looked at Nagios and saw an infinite community of support/plugins. I have read much complaining about the tool. Meanwhile, I am also seeing it being heavily used. Religious war? Random internet whining?

Constraints for my choices are:

1. Mix of various flavours of *ix and winblows (server and desktop) on the vlans.
2. Mix of IPMI, SNMP, SQL Server, Apache, IIS, Exchange, S4B Server, other common servers and a couple of custom apps.
3. Maybe a little bit of integration to Azure and AWS (stretch) but I have VMs in both.
- some level of encryption if over the WAN.
4. Mostly looking for notification rather than fancy graphing or stats.
5. These are primarily dev and test configs and can be brought up and down without significant business impact in most cases.

Adding a tool to my resume isn't terribly important at this point in my career but I prefer to be able to talk with other folk and have them understand what I am talking about so I don't want to pick up a great tool that no-one knows. BUT if Nagios actually blows I'd like to know in advance before I invest much time into Nagios Core.

Thanks,
R

 

[cheezehead]

Nagios works...for a homelab I think it's a bit heavy of a time sink. I ended up switching over to LibreNMS awhile back because it wasn't a time sink..it doesn't have the bells & whistles for Exchange/IIS/SQL but otherwise works well for everything else.

 

[abundantmussel]

Ive ran nagios since 2012 but switched to its fork Icinga in late 2013. Today I run sensu instead of both (it can also use nagios plugins which is handy).


Sent from my iPhone using Tapatalk

 

[itguy82]

We run Nagios XI work. It's a POS pile of software, but there are not a lot of alternatives for our scale unless you want to spend $$$$$$ for something like Solarwinds. Zabbix might work for some. We decided against Zabbix for obscure/idiosyncratic reasons.

 

[wildchild]

Take a look at zabbix, comming from nagios you'll find it releaving.
Add to that the numerous premade templates and low level discovery, and your golden.
In centos and ubuntu it's in the standard repo's so installing and upgrading is a breeze

 

[cheezehead]

... $$$$$$ for something like Solarwinds....

For a homelab, Solarwinds does have a thing with OpGenie which has a free version (pay attention to the limits) SolarWinds Alert Central - A Strategic Partnership with OpsGenie

For business they can be expensive, but depending upon scale and time of year they do offer discounts...at least on the initial licensing, I haven't heard of them discounting the ongoing support contracts much.

 

[markarr]

I like both PRTG (limit of 100 sensors) and observium. PRTG needs to be on windows but can monitor pretty much anything, I use it for the build integrations and then observium for some linux and network. Neither are very hard to setup so it is nice for home labs so you can work on stuff rather than setting up monitoring.

 

[mstone]

+1 for sensu. Wasn't much of a fan of zabbix.

 

[BLinux]

I've used Nagios for a long time, since its beginnings as "netsaint". I did a massive deployment about 10 yrs ago for Sony and have written several custom plugins for it. Once you get use to it, and develop some practices to manage it cleanly, it works. Monitoring isn't my specialty, so I haven't touched it much since then except for a few occasions here and there. I don't know how much it has improved since... but because it has been around for a long time, it's accumulated years of community contributions - that's probably what you're seeing; take a look at the dates for some of those plugins.

However, these days, I hear people talking about a bunch of newer tools. I don't hear Nagios get mentioned much anymore and I certainly hear about Zabbix a lot more. On the surface of things, Zabbix looks prettier, but I don't know much else about it except that it is widely used today with one of my clients; but I don't dabble in that area for them.

As for "relevance", I don't suspect Nagios would be the best choice. I still think it's a very useful tool. But there are ever more variety of tools in this space now than before, so I suspect the older tools (inc. Nagios) have always lacked something, or people/organization's needs are evolving, or both and new tools have stepped in to solve those problems. So, I would probably focus on those other tools if you want to get familiar with something you hope will be more relevant. However, like i said, there are so many monitoring/trending tools out there now and every where I go, people have their different preferences - if I ever did consulting in that space again, I would really have to zero in on a specific use-case and see what best meets the requirement. Nagios would be my "generic" backup solution to go to if I couldn't find something better.

Over the course of a couple decades, I've used a variety of monitoring tools, open source and commercial. The thing that I've gathered, and I suspect it still applies today even with the newer tools, is that they all require some degree of customization, plugin writing, integration work with other tools, etc. to get it to a state that works well for any particular environment. Unless you have a really simple environment (which you may, idk), I haven't seen a tool in this area that is "intelligent" and auto-configures itself and doesn't require some labor from someone who specializes in that tool.