1

We have around 50 virtual machines. I installed collectd, graphite and grafana and we have some traffic statistics like these:

That produces a graph like this:

The lines are incoming and outgoing bandwidth in KB/sec, but I would like to have statistics that can tell me how much traffic is used in GB's between 09:00 and 17:00 on the entire server farm.

What would be a good way to find out how much data has been users over a group of servers?

4
  • 1
    Monitor the switch/router ports with your uplink?
    – HBruijn
    Jun 27, 2016 at 10:29
  • 1
    It's a cloud-like environment, I don't really have access to the switches/routers.
    – ujjain
    Jun 27, 2016 at 11:30
  • I guess that's supposed to be the beauty of the cloud. But in the physical world, this information would be trivial to acquire. Do you have access to the cloud environment's API or metering tools?
    – ewwhite
    Jun 27, 2016 at 12:20
  • Not really, it's part an OpenStack environment and partially they are physical servers.
    – ujjain
    Jun 28, 2016 at 6:27

2 Answers 2

0

Using an integral and scaling by 60 seemed to do the trick.

0

Have you tried using Netflow monitoring?

You could monitor the IP traffic by using the NetFlow protocol. This protocol allows to you review the traffic's most useful patterns and general data.

'NetFlow' is a network protocol, developed by Cisco Systems to collect IP traffic information. It has become an industrial standard for network traffic monitoring and is currently supported by several platforms besides Cisco's IOS and NXOS like Juniper devices, Enterasys Switches and operating systems like Linux, FreeBSD, NetBSD and OpenBSD.

If a NetFlow capable router is not available, but you use a Linux server to route your traffic, you may install a NetFlow software probe which sends all NetFlow-related information to its server.

In Linux there is a program called fprobe which obtains the traffic and sends it to a NetFlow Server. By this program you're able to generate NetFlow traffic which goes through its interfaces, e.g.:

/usr/sbin/fprobe -ieth0 -fip 192.168.1.185:9995

Once the traffic has been generated, you're able to review the traffic's statistics by entering the following command:

nfdump -R /home/netflow_data/

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .