Monitoring Network traffic

[marsdenprimary]

Hello. I have a bit of an issue that needs pinpointing and I want to know if there is any free software that lets me monitor traffic of individual machines on our network? Thanks in advance.

[round2it]

solar winds have some tools think they do a 30 day trial

[FN-GM]

Most managed switches have SNMP and NetFlow built into them.

[networkmangler]

I use cacti to monitor network traffic https://www.cacti.net/

[aguacate]

For a small network, i would use Passler's PRTG which is free and runs on windows, if you are happy to play with Linux i would go for LibreNMS , both free up to a point, but require snmp configuration on end devices

 

A

[cpjitservices]

We use Zabbix and PRTG Free (100 sensors)

[marsdenprimary]

Thanks everyone I'm going to use Cacti for now

[epoch_roots]

Zabbix is the most powerful monitoring solution around in my opinion. The ability to automatically raise an alert and then attempt mitigations on servers is a real winner for me.

Service down? Run a powershell script. Doesn't fix it? Send an email/SMS/post to Slack, and kick off failover services. The possibilities are endless when triggers can be used to execute applications based on reported conditions. The screens are a great way to get an overview of your service offering too. Really powerful stuff but likely overkill for the types of environments we have. Due to my familiarity with it I have installed it here for monitoring around 20 servers. In a previous role we were using it to monitor 20,000+ items, with the server receiving 150 metrics a second.

[markwilfan]

Zabbix is the most powerful monitoring solution around in my opinion. The ability to automatically raise an alert and then attempt mitigations on servers is a real winner for me.

Service down? Run a powershell script. Doesn't fix it? Send an email/SMS/post to Slack, and kick off failover services. The possibilities are endless when triggers can be used to execute applications based on reported conditions. The screens are a great way to get an overview of your service offering too. Really powerful stuff but likely overkill for the types of environments we have. Due to my familiarity with it I have installed it here for monitoring around 20 servers. In a previous role we were using it to monitor 20,000+ items, with the server receiving 150 metrics a second.

We use zabbix too. Nice bit of kit for free

[synaesthesia]

The Dude is fairly simple and surprisingly powerful for this tool

[cpjitservices]

Zabbix is the most powerful monitoring solution around in my opinion. The ability to automatically raise an alert and then attempt mitigations on servers is a real winner for me.

Service down? Run a powershell script. Doesn't fix it? Send an email/SMS/post to Slack, and kick off failover services. The possibilities are endless when triggers can be used to execute applications based on reported conditions. The screens are a great way to get an overview of your service offering too. Really powerful stuff but likely overkill for the types of environments we have. Due to my familiarity with it I have installed it here for monitoring around 20 servers. In a previous role we were using it to monitor 20,000+ items, with the server receiving 150 metrics a second.

 

I havent gone that in depth with it but my next experiment will be monitoring some remote servers over the internet. Only problem I see with it is my other colleagues aren;t going to want to edit config files like I think you have to do to specify port numbers etc (active server addresses etc)

 

There is the Zabbix Agent with the config file.

[epoch_roots]

It's not that difficult really. editing the zabbix agent config on a server by server basis is pretty rare, and you can use a single "template" config for deployment to all servers without having to edit anything else.

I wrote a powershell script that would automatically deploy the agent to all servers in an AD domain but didn't take it with me when I left the role!

[cpjitservices]

It's not that difficult really. editing the zabbix agent config on a server by server basis is pretty rare, and you can use a single "template" config for deployment to all servers without having to edit anything else.

I wrote a powershell script that would automatically deploy the agent to all servers in an AD domain but didn't take it with me when I left the role!

 

OK thanks. I'll see if I can create a template for it.

[Mistert89]

Paessler tracked down our problem

[cpjitservices]

It's not that difficult really. editing the zabbix agent config on a server by server basis is pretty rare, and you can use a single "template" config for deployment to all servers without having to edit anything else.

I wrote a powershell script that would automatically deploy the agent to all servers in an AD domain but didn't take it with me when I left the role!

 

Just to add, we do have to edit the config file quite often as we take up monitoring servers for companies over the internet that are not part our AD.

Edited October 27, 2017 by cpjitservices

[epoch_roots]

Just to add, we do have to edit the config file quite often as we take up monitoring servers for companies over the internet that are not part our AD.

A zabbix agent config only needs like 3 lines - the server address, a hostname, and a logfile location. I think thats all anyway. In terms of what items are checked, these can be pushed/pulled by the server. The agent config is used for setting up commands/userparams specfic to that server/service.

You can also use autodiscovery within zabbix to automatically create & start monitoring the host once the agent is deployed to it.

You can deploy pretty quick by zipping the files up & including a zabbix_install.bat with something like the following in it:

 

echo Installing Zabbix agent service
zabbix_agentd.exe --config "C:\zabbix\zabbix_agentd.conf" --install
echo Starting Zabbix agent
net start "Zabbix agent"

Edited October 27, 2017 by epoch_roots
incl quote

[cpjitservices]

A zabbix agent config only needs like 3 lines - the server address, a hostname, and a logfile location. I think thats all anyway. In terms of what items are checked, these can be pushed/pulled by the server. The agent config is used for setting up commands/userparams specfic to that server/service.

You can also use autodiscovery within zabbix to automatically create & start monitoring the host once the agent is deployed to it.

You can deploy pretty quick by zipping the files up & including a zabbix_install.bat with something like the following in it:

 

echo Installing Zabbix agent service
zabbix_agentd.exe --config "C:\zabbix\zabbix_agentd.conf" --install
echo Starting Zabbix agent
net start "Zabbix agent"

 

Thanks man, this is similar to what I had setup previously, though the Autodiscover didnt seem to work very well. DO I need to configure anything for it to work? ports opening perhaps? As the servers are moinitored over the internet.

[epoch_roots]

According to the docs its as simple as creating an action for the Auto Registration event but I never remember using it. 99% of our servers were LAN and autodiscovery was via SNMP (although we used low level discovery extensively for item creation). It doesn't require any additional ports opening, just what you need for active agent checks.

There's some decent vmware / hyperv autodiscovery that will add all guests on a host in, but thats something entirely different.

[cpjitservices]

According to the docs its as simple as creating an action for the Auto Registration event but I never remember using it. 99% of our servers were LAN and autodiscovery was via SNMP (although we used low level discovery extensively for item creation). It doesn't require any additional ports opening, just what you need for active agent checks.

There's some decent vmware / hyperv autodiscovery that will add all guests on a host in, but thats something entirely different.

 

Thanks mate, I'll give it another look - if I get stuck I'll come back to you if that is ok? and likewisae if there is anything I can assist with just PM me.

[epoch_roots]

Sure, no problem at all. The zabbix docs can sometimes be really confusing (I'm guessing due to support/maintenance agreements being their only revenue stream) so PM me if you get stuck with anything.