Solarwinds Versus Nagios

Hi,

Has anyone had any experience of working with Solarwinds and Nagios. I have used Solarwinds and think this is an excellent network monitoring tool. I wondered what the pro's and cons were of both and which one you preferred.

I would be interested in hearing from anyone who has an opinion on this.

Thanks CJ

Find more posts tagged with

Comments

W Stewart

I've used them both although I haven't messed with nagios as much as solarwinds. Solarwinds had some nice features for monitoring and I learned a lot about it in the short time that I actually worked with it. Nagios just seems to work and when a server is down, I investigate. Solarwinds seemed to have more issues but that could have been because it was running on a server 2003 box and possibly old hardware where as the company I'm working at now runs nagios on a linux box with some decent hardware. Again, I haven't delved deep into Nagios and it's possible that what I'm looking at is just a webpage front end to nagios that not everybody uses but it's still pretty nice regardless.

Not sure if different environments have different prefferences but the company I was working at that used solarwinds was an ISP. The company that uses nagios is a web hosting company. I've also seen a linux admin at a previous job use nagios so it may be that nagios is more popular among linux if not servers altogether.

cjthedj45

W Stewart wrote: »

I've used them both although I haven't messed with nagios as much as solarwinds. Solarwinds had some nice features for monitoring and I learned a lot about it in the short time that I actually worked with it. Nagios just seems to work and when a server is down, I investigate. Solarwinds seemed to have more issues but that could have been because it was running on a server 2003 box and possibly old hardware where as the company I'm working at now runs nagios on a linux box with some decent hardware. Again, I haven't delved deep into Nagios and it's possible that what I'm looking at is just a webpage front end to nagios that not everybody uses but it's still pretty nice regardless.

Not sure if different environments have different prefferences but the company I was working at that used solarwinds was an ISP. The company that uses nagios is a web hosting company. I've also seen a linux admin at a previous job use nagios so it may be that nagios is more popular among linux if not servers altogether.

Hi thanks for the reply. I get the impressions that the Linux guys will perfer Nagios but it this because it perhaps offers really good features to monitor Linux servers. I'm not sure how it would perform as a Network Monitor. I think Solarwinds is best for network guys wanted to monitor the network and Nagios for monitoring Linux servers. I maybe wrong as I have not used Nagios.

Hypntick

It really depends on what you want to do as far as network monitoring. Where I was previously used Nagios as mostly an up/down monitor for our clients connections. Really didn't get more in depth than that, but it worked pretty well for it.

GAngel

On the nagios end it's a very in depth product. We were monitoring down to cpu cycles with no issues. It's a PITA to setup but once it's running its low maintenance.

Solarwinds seems a bit more user friendly but I havn't really used it.

undomiel

If you've got the budget and a lower learning curve then I would say go for SolarWinds. Otherwise I'd opt for Nagios. The learning curve is definitely sharper and you'll need to get used to text file editing and I would definitely recommend getting very friendly with sed and awk. The plugins and scripting is where Nagios can really shine for getting into more granular detail on what and how you're monitoring.

ChooseLife

cjthedj45 wrote: »

I get the impressions that the Linux guys will perfer Nagios but it this because it perhaps offers really good features to monitor Linux servers.

True dat (Nagios fan and Linux admin here

)

netsysllc

This is a nagios fork that has more options and is simpler setup and configuration Shinken | The next Industry Standard in IT Monitoring

ChooseLife

netsysllc wrote: »

This is a nagios fork that has more options and is simpler setup and configuration Shinken | The next Industry Standard in IT Monitoring

Good stuff!
Also, OpsView is similar to that

cjthedj45

Thanks Guys appreciate the comments. Are any of you guys network admins though or are you more systems/linux. I have had a quick look round and I'm not sure if Nagios really peforms as well as Solarwinds as a network monitor tool. Solarwinds was built as a network monitor and I think Nagios was built to monitor servers. Therefore can you really get as good monitoring from Nagios for the network. I like Solarwinds becuase of these points:

Automatic network discovery of devices.
Health Monitoring and alerting of all network equipment (I remember once Solarwinds flagged sometype of memory leak on a voip gateway if left unattended this could of caused a big issue)
good reporting stats such as network availabilty, perfomance graphs etc
Solarwinds allows you to import interactive network maps
Bandwidth and utilization reports (This can be a god send when people are telling you the network is slow. I can take a quick look at the network links and ensure there is no network saturation)
Solarwinds also has netflow capabilty not sure if Nagios does this?

This Shinken sounds interesting. Has anyone used or is using it? Whats it like? I just tried to see a demo on ther site but it did not work.

thegoodbye

I prefer hobbit(now xymon) and it's free

f0rgiv3n

I have setup a Nagios server from scratch as well as worked with Solarwinds pretty extensively. From my perspective they are on two completely different playing fields. Nagios definitely has its place, it's free... and it works well in a smaller environment. Solarwinds is expensive but it is a lot more robust than Nagios. Solarwinds does require you to install "Modules" in order to have in depth application monitoring, etc... Then again, so does Nagios... but you have to pay an arm and a leg for Solarwinds.

So depending on how big your environment is, you'll have to evaluate if the cost is worth it. Nagios, you'll spend your money you save on time to set it up. It takes a lot of time and determination to understand its inner-workings.

Solarwinds is a lot more than just a network monitoring tool. A quick example: You can develop "ghost runs" of an application and have it monitor the latency between steps. Meaning, you could configure it to load a web page, login to the webpage and run a link to gather data, all the while timing how long it takes to get from step to step. That gives you an idea of how much more Solarwinds has to it.

Nagios does have many open-source modules you can use (hell I even used one to telnet into an old AS400 and monitoring running processes).

So like I said, it depends on the environment and what you want out of the system. To answer the question about netflow, Nagios itself I don't think can do netflow but it can pair up with another module that can (and you still get to see it from a single pane of glass). Any specific questions let me know!

There's a ton of open source software out there that use Nagios and not. Ninja (front end GUI for nagios), Zenoss, What's Up Gold (YUCK!), etc... You could also get things like Alienvault (nagios is built in) that has more than just monitoring in it (it's an Open Source IDS). Cacti can be paired with Nagios to provide you with graphs for bandwidth utilization... Ok now I'm starting to blab, I'll end it here.

Mrock4

I have pretty extensive experience deploying SolarWinds Orion in large environments (5,000+ nodes). No experience with Nagios. SolarWinds Orion seems to be either loved, or hated. In my experience, the customers who hated it, also had it poorly deployed (ie: using an old database server that's bogged down with various services as well as handling the solarwinds database)- or otherwise not setup properly.

At the end of the day, if you deploy it right, Orion can literally do almost anything. But that's a call you (or your organization) have to make- do you have the resources to put into learning/deploying it (or hiring someone to do it)? If the answer is "no" - then SolarWinds will fail. Well..anything will fail if you can't put the appropriate time into it.

I will say, we did some pretty slick things with SolarWinds custom reporting/alerts, and I really have enjoyed working with it.

lsud00d

I've worked extensively with Nagios but also evaluated Solarwinds and it is used in house for production. As stated they can do the same general things but they specialize in different aspects.

As a sys admin I really like SW transactions module for assuring web app availability and response. Network really likes the flow monitoring for our fabric. It has great drill downs for conversations.

But, network team still uses our Nagios/Cacti/Weathermap setup for bandwidth utilization monitoring which I have not seen SW do on a WAN level. I feel the snmp monitoring is easier in Nagios/Cacti....the interface polling is more straightforward to customize.

SW is easy out the box and hands off. You can set subnets to crawl for network dependency configuration which is a process in Nagios. The interface is clean and can provide great at a glance dashboard functionality for a NOC monkey.

Nagios is beautiful in its archaic GUI and flat text configs however there is more freedom if you are down with linux which is why I prefer it to SW, not to mention the price comparison.

mokeefe

Nagios XI is something that should be considered as it's Nagios with out the need to enter the config files. It can do all servers and Network devices very well. You can contact Nagios Ent for more information and a free fully functional 60 day trial. It worth at least looking into before spending money. I am biased as I work at Nagios and hear a lot about Solarwinds from current customers, FYI. Let me know if I can help at all

ITtech2010

Can any of you guys give me good tips on how you configured the latest network perfomance monitor for solarwinds? I tried working with support but the charts still is not displaying correctly! When i click on a node to monitor the links the interfaces are at 0-1%

Mrock4

A few things to check if the data isn't being displayed properly in solarwinds:

1) Ensure the database is sync'ing properly to the front end..go to the admin tab (top right)..then I believe it's under polling engine details..it shouldn't be longer than a few seconds since it last sync'd. I've seen a few installs that were at 48+ hours..rebooted the front end (not the DB), and it corrected the issue.

2) Make sure the device itself is configured correctly. An example is if the BW statements have been altered on the device for traffic engineering purposes (such as with EIGRP), it won't look right in SolarWinds.

3) SolarWinds best practice is to NOT place the DB and front end on the same server..I've done it, but it's something to keep in mind. If the DB gets bogged down you'll get weird results.

Other then that it's pretty straightforward. What version of Orion core are you using?

Cherper

lsud00d wrote: »

But, network team still uses our Nagios/Cacti/Weathermap setup for bandwidth utilization monitoring which I have not seen SW do on a WAN level. I feel the snmp monitoring is easier in Nagios/Cacti....the interface polling is more straightforward to customize.

We use SolarWinds to monitor bandwidth utilization. We replaced Nagios and CA's eHealth with SW and wouldn't ever go back. We monitor only routers and switches (3500) and our units servers (10 servers). The unit that hosts all of the other servers and VMs (approximately 2500) still use Nagios, and they hate it. They have to have a full time employee to deal with Nagios.

xkilian

The original question is Nagios versus Solarwinds. But as Shinken is a Nagios compatible python rewrite, I think it fits.

Out of the box functionality is Solarwinds hands down, it is commercial software and it shows. But..

Here is why Shinken may be an interesting alternative for some groups of users for Network monitoring.

Shinken is built for scalability
Shinken has a native SNMP module which supports SNMPv2c GetBulk
Shinken has support (and getting better) for using Graphite as a back-end time-series database and visualization engine
Shinken can use genDevConfig to discover and create targets automatically for typical network devices
Graphite has very amazing scalability and performance, do not try to scale the Solarwinds database to the same level as Graphite! Relational database versus time-series database is no comparison.

Shinken is not meant to be used directly as a trap receiver. You should use your Nortel/Avaya VPFM console for that, or Cisco Prime, or Juniper mgmt to process the traps and forward them as alerts into Shinken. Or.. I personally find that OSSEC is a great syslog receiver (syslog always receives same as traps) and it can forward any interesting alerts to Nagios/Shinken. Rule building in OSSEC dead simple too.

Shinken does not have great built in dashboards for specific network reports. But graphite does help.

Shinken does not have auto-discovery of networks and dynamic map building. It is more typically used with an existing CMDB.

Hope this helps

xkilian

mass175

We were using SolarWinds and Nagios and replaced them with SevOne. We used them to monitor our network, server, F5, Avaya VoIP, applications(JMX, HTTP, WMI, DNS), DBs and for capacity planning (6X). They give you 20 different polling technologies out of the box that are included with pricing, Crazy!!!!

In my previous job we used Nimsoft, Cacti and VitalSuite and they also replaced with SevOne.

IF you are an advanced user, I would strongly recommend SevOne.

Why SevOne? Performance Visibility. Solarwinds vs. SevOne: Network Performance Monitors Compared

Good Luck!

joeshestak

I have not used Nagios, but Ive used both solarwinds and MindArray IPM and both are great products, I just personally think MindArray IPM is a better value.

Joe_Sagona

Solarwinds is a three pronged subscription based service with NPM, APM and Database performance. The GUI will be much cleaner than Nagios and it will be easier to config. Additionally, you will get the extra visuals and reports. I guess you get what you pay for. Nagios is open source and will give you pretty much everything you need. Just need to be aware that the configs are not as straight forward.

For simple NPM, I would go with Nagios as it has a large ecosystem of plugins and integrations, which are worth the extra money. If you are running Nagios on a larger network with multiple tiers, I would look into PagerDuty and BigPanda. PagerDuty will allow you to create escalation rules to help route the tickets to the right people in the appropriate tiers. BigPanda will allow you to optimize your resources by compressing the alerts you get from Nagios into manageable incidents and reducing your time to resolution.