Sys Admins -- Any good network/system monitoring software?

GLaDOS11

I'm looking to install some kind of network/server monitoring software on our network. I'm the only member of the IT department and we have quite a few servers. It's getting harder and harder to manage them all by myself. What kind of software do you guys use to monitor your network, let you know when servers go down, excessive errors in event viewer, etc? Ideally, this would be free or low cost, but I'm open to all suggestions.

I'm trying to find out what other companies use and how they approach this problem. Thanks!!

shochan

We use SolarWinds, but it is a working progress...lots of bugs & it's not free.

SeekBytes

I think that those two are the most widely adopted:

https://www.nagios.org/
Cacti® - The Complete RRDTool-based Graphing Solution

alias454

Another couple of alternatives
https://www.icinga.org/
http://www.shinken-monitoring.org/
http://www.zenoss.org/

4_lom

We use System Center Operations Manager. Extremely powerful monitoring/automation software from Microsoft. Somewhat expensive. Also not the most user friendly. It's fairly simple to setup and get it running. However, if you want to customize/add monitoring or automation, you may need to spend a solid day or two readying technet articles or asking others for help.

Another good solution is that I've used in the past is WhatsUp Gold. It's expensive though. Can't remember the exact cost, but I remember getting sticker shock when I saw the first bill lol.

I've also used LabTech, N-Able, and Level Platforms in a previous lifetime. LabTech is extremely powerful (probably the most powerful monitoring/automation platform I have ever experienced). However, it is somewhat buggy and unpredictable at times (not a good combination btw).

4_lom

If you're looking free software, nothing beats Nagios IMO

dhay13

Nagios and it is free

beads

Nagios is so limited its not funny. Check out ExtraHop for a expensive tool but worthwhile beyond compare. There is a community edition but it doesn't get you anywhere near what the paid subscription does.

- b/eads

iBrokeIT

Zenoss would be the way to go if free is a hard requirement

Ertaz

I've used Solarwinds, Nagios, PRTG, and Osisoft PI. The first three were OK, but very constrained by the canned UI and notification systems. PI was just a data collection framework with an SNMP interface. It was a complex initial config but it allowed us to make decisions based on comparative performance equations and the deep dive analytics tools it had.

GLaDOS11

Thanks very much for all of the replies. I'm going to check some of these out this afternoon. Looks like there are a lot of good options. I'll report back if I have any more questions.

Thanks again everyone

LDhiker

Please post back what you have found.

Currently we use solarwinds, looking for another option that isn't so expensive.
In the last month, I have tried Netgrunch and intermapper. Looking at trying out PRTG after the intermapper trial is over.

mdhisapro

My previous employer used WhatsUp Gold by IPswitch. Not free; but has a lot of functionality. Allowed us to create maps, and overlay devices on the maps, restart failed services, monitor services, processes, CPU usage as well as storage etc and notify when actions like that took place.

They also used the Whats Virtual plugin, to monitor their virtual presence.

alias454

LDhiker wrote: »

Please post back what you have found.

Currently we use solarwinds, looking for another option that isn't so expensive.
In the last month, I have tried Netgrunch and intermapper. Looking at trying out PRTG after the intermapper trial is over.

We use PRTG and I find it to be pretty decent to use. it has some drawbacks but overall it is a good product. let me know if you have any questions.

ColeAltDelete

What operating system(s) are these servers running? Windows/Linux/Unix - a mix?
What do you want to monitor? Server health, disk space, performance, services, something more advanced?
Are these systems virtualized? Are you also looking to monitor your storage? (SAN/NAS/etc)

We use Solarwinds for our environment - it's fairly robust but certain things feel more difficult to do than using other monitoring products. I've heard excellent things about PRTG - however, I think it's sensor based so that could make it expensive fairly quickly. Nagios is usually the go to for an open-source product. Depending on what you're monitoring - sometimes you may need multiple monitoring tools to accomplish proper monitoring. (i.e. vCOPS for VMware for instance)

alias454

PRTG will allow 100 sensors for free with full functionality I think. I don't know what the OP consider "Quite a few server" so it might be enough for his use case. https://www.paessler.com/download/prtg-download

The upfront cost can be a lot but the yearly maintenance is like 20% on going. We put 3 sensors on each server by default CPU, RAM, and Disk Space. Outside of that, we add things as needed.

GLaDOS11

ColeAltDelete wrote: »

What operating system(s) are these servers running? Windows/Linux/Unix - a mix?
What do you want to monitor? Server health, disk space, performance, services, something more advanced?
Are these systems virtualized? Are you also looking to monitor your storage? (SAN/NAS/etc)

We use Solarwinds for our environment - it's fairly robust but certain things feel more difficult to do than using other monitoring products. I've heard excellent things about PRTG - however, I think it's sensor based so that could make it expensive fairly quickly. Nagios is usually the go to for an open-source product. Depending on what you're monitoring - sometimes you may need multiple monitoring tools to accomplish proper monitoring. (i.e. vCOPS for VMware for instance)

They are all running some variation of Windows Server from 2003 - 2016. Many of them are virtualized. I'm looking to monitor exactly what you mentioned: server health, disk space, performance, Windows Event Logs, Windows Update, etc. We do not have any SAN or NAS that we need to monitor.

I'm going to take a look at SCOM first because I don't know if we have access to that through Microsoft Partner Network. If we do that might be good. I'll also look at some of the others mentioned if SCOM doesn't work out.

Lexluethar

IF you want something easy to configure and relatively cheap check out ManageEngine OPManager. It does a good job of monitoring switches, it can monitor specific ports (ingress and egress) as well as the ability to monitor virtual and physical severs. There is no agent installation which is also a plus.

It's not an enterprise level application IMO - support is sub par but the product itself is really easy to configure, install and manage. It's also cheap compared to competitors (there are cheaper solutions but none that are as easy to configure). It's the perfect mix for small IT shops.

Kinet1c

Plenty of options to be found at this site: IT Landscape for sysadmins

MAC_Addy

Nagios is pretty awesome. However, I really like PRTG. I have also used SolarWinds in the past, but it is expensive. You get what you pay for I suppose.

gkca

Zabbix http://www.zabbix.com/

Verities

gkca wrote: »

Zabbix http://www.zabbix.com/

+1 for Zabbix. Better than Nagios in my opinion and its free. It integrates well with Grafana and you can configure it to use SNMPv2 or v3 (added security) and you can literally monitor anything with it. The learning curve is considerable when figuring out specific mibs, but you can use the default templates as a baseline.

Mike7

Another +1 for Zabbix especially if you are comfortable with Linux and $$$ is a priority. The SNMP MIBs learning part is a challenge but there are custom templates for various network devices around. For Windows boxes, use the Zabbix agent instead of SNMP as it supports perfmon counters and allows custom actions. We had the agent restart a NT service that was leaking memory whenever RAM usage hit a threshold. We also configured Zabbix to send custom SMS alerts and even phone calls via php script that calls Nexmo API. You can also find mobile Zabbix apps most of which are free.

FWIW, the CentOS team uses Zabbix.

Verities

Mike7 wrote: »

Another +1 for Zabbix especially if you are comfortable with Linux and $$$ is a priority. The SNMP MIBs learning part is a challenge but there are custom templates for various network devices around. For Windows boxes, use the Zabbix agent instead of SNMP as it supports perfmon counters and allows custom actions. We had the agent restart a NT service that was leaking memory whenever RAM usage hit a threshold. We also configured Zabbix to send custom SMS alerts and even phone calls via php script that calls Nexmo API. You can also find mobile Zabbix apps most of which are free.

FWIW, the CentOS team uses Zabbix.

From a Linux perspective (I don't use it for Windows), you can do custom actions without the Zabbix agent. You can also have a service account to run the service restarts in the event a service goes down and you need to get it back up. Importing templates is another alternative but you have to tweak them to get the checks to perform within the specific time slots you choose. I had to tune down default templates from 5 minutes to 30 seconds because that long is simply unacceptable for a service to be down. I do enjoy the flexibility of the notifications; I normally use internal smtp for email alerts, but if our mail server goes down Zabbix will use external smtp (gmail) to send me an alert at an alternate email.