Peter

Author Archives: Peter

Large flow marking using BGP Flowspec

Elephant Detection in Virtual Switches & Mitigation in Hardware discusses a VMware and Cumulus demonstration, Elephants and Mice, in which the virtual switch on a host detects and marks large "Elephant" flows and the hardware switch enforces priority queueing to prevent Elephant flows from adversely affecting latency of small "Mice" flows.

SDN and WAN optimization describes a presentation by Amin Vahdat describing Google's SDN based wide area network traffic engineering solution in which traffic prioritization allows Google to reduce costs by fully utilizing WAN bandwidth.

Deconstructing Datacenter Packet Transport describes how priority marking of packets associated with large flows can improve completion times for flows crossing the data center fabric. Simulation results presented in the paper show that prioritization of short flows over large flows can significantly improve throughput (reducing flow completion times by a factor of 5 or more at high loads).

This article demonstrates a self contained real-time Elephant flow marking solution that leverages the real-time visibility and control features available using commodity switch hardware.

The diagram shows the elements of the solution. An instance of the sFlow-RT real-time analytics engine receives streaming sFlow telemetry from a pair of edge routers. A mix of many small flows mixed Continue reading

Real-time network and system metrics as a service

The sFlow-RT real-time analytics engine receives industry standard sFlow telemetry as a continuous stream from network and host devices and coverts the raw data into useful measurements that can be be queried through a REST API. A single sFlow-RT instance can monitor the entire data center, providing a comprehensive view of performance, not just of the individual components, but of the data center as a whole.

This article is an interactive tutorial intended to familiarize the reader with the REST API. The examples can be run on a laptop using recorded data so that access to a live network is not required.

The data was captured from the leaf and spine test network shown above (described in Fabric View).
curl -O https://raw.githubusercontent.com/sflow-rt/fabric-view/master/demo/ecmp.pcap
First, download the captured sFlow data.

You will need to have a system with Java or Docker to run the sFlow-RT software.
curl -O https://inmon.com/products/sFlow-RT/sflow-rt.tar.gz
tar -xzf sflow-rt.tar.gz
./sflow-rt/get-app.sh sflow-rt browse-metrics
./sflow-rt/get-app.sh sflow-rt browse-flows
./sflow-rt/get-app.sh sflow-rt prometheus
./sflow-rt/start.sh -Dsflow.file=$PWD/ecmp.pcap
The above commands download and run sFlow-RT, with browse-metrics, browse-flows, and prometheus applications on a system with Java 1.8+ installed.
docker  Continue reading

NVIDIA, Mellanox, and Cumulus

Recent press releases, Riding a Cloud: NVIDIA Acquires Network-Software Trailblazer Cumulus and NVIDIA Completes Acquisition of Mellanox, Creating Major Force Driving Next-Gen Data Centers, describe NVIDIA's moves to provide high speed data center networks to connect compute clusters that use of their GPUs to accelerate big data workloads, including: deep learning, climate modeling, animation, data visualization, physics, molecular dynamics etc.

Real-time visibility into compute, network, and GPU infrastructure is required manage and optimize the unified infrastructure. This article explores how the industry standard sFlow technology supported by all three vendors can deliver comprehensive visibility.

Cumulus Linux simplifies operations, providing the same operating system, Linux, that runs on the servers. Cumulus Networks and Mellanox have a long history of working with the Linux community to integrate support for switches. The latest Linux kernels now include native support for network ASICs, seamlessly integrating with standard Linux routing (FRR, Quagga, Bird, etc), configuration (Puppet, Chef, Ansible, etc) and monitoring (collectd, netstat, top, etc) tools.

Linux 4.11 kernel extends packet sampling support describes enhancements to the Linux kernel to support industry standard sFlow instrumentation in network ASICs. Cumulus Linux and Mellanox both support the new Linux APIs. Cumulus Linux uses the open source Continue reading

Monitoring DDoS mitigation

Real-time DDoS mitigation using BGP RTBH and FlowSpec and Pushing BGP Flowspec rules to multiple routers describe how to deploy the ddos-protect application. This article focuses on how to monitor DDoS activity and control actions.

The diagram shows the elements of the solution. Routers stream standard sFlow telemetry to an instance of the sFlow-RT real-time analytics engine running the ddos-protect application. The instant a DDoS attack is detected, RTBH and / or Flowspec actions are pushed via BGP to the routers to mitigate the attack. Key metrics are published using the Prometheus exporter format over HTTP and events are sent using the standard syslog protocol.
The sFlow-RT DDoS Protect dashboard, shown above, makes use of the Prometheus time series database and the Grafana metrics visualization tool to track DDoS attack mitigation actions.
The sFlow-RT Countries and Networks dashboard, shown above, breaks down traffic by origin network and country to provide an indication of the source of attacks.  Flow metrics with Prometheus and Grafana describes how to build additional dashboards to provide additional insight into network traffic.
In this example, syslog events are directed to an Elasticsearch, Logstash, and Kibana (ELK) stack where they are archived, queried, and analyzed. Grafana Continue reading

Pushing BGP Flowspec rules to multiple routers

Real-time DDoS mitigation using BGP RTBH and Flowspec describes the open source DDoS Protect application. The software runs on the sFlow-RT real-time analytics engine, which receives industry standard sFlow telemetry from routers and pushes controls using BGP. A recent enhancement to the application pushes controls to multiple routers in order to protect networks with redundant edge routers.
ddos_protect.router=10.0.0.96,10.0.0.97
Configuring multiple BGP connections is simple, the ddos_protect.router configuration option has been extended to accept a comma separated list of IP addresses for the routers that will be connecting to the controller.
Alternatively, a BGP Flowspec/RTBH reflector can be used to propagate the controls. Flowspec is a recent addition to open source BGP software, FRR and Bird, and it should be possible to use this software to reflect Flowspec controls. A reflector can be a useful place to implement policies that direct controls to specific enforcement devices.

Support for multiple BGP connections in the DDoS Protect application reduces the complexity of simple deployments by removing the requirement for a reflector. Controls are pushed to all devices, but differentiated policies can still be implemented by configuring each device's response to controls.

Kubernetes testbed

The sFlow-RT real-time analytics platform receives a continuous telemetry stream from sFlow Agents embedded in network devices, hosts and applications and converts the raw measurements into actionable metrics, accessible through open APIs, see Writing Applications.

Application development is greatly simplified if you can emulate the infrastructure you want to monitor on your development machine. Docker testbed describes a simple way to develop sFlow based visibility solutions. This article describes how to build a Kubernetes testbed to develop and test configurations before deploying solutions into production.
Docker Desktop provides a convenient way to set up a single node Kubernetes cluster, just select the Enable Kubernetes setting and click on Apply & Restart.

Create the following sflow-rt.yml file:
apiVersion: v1
kind: Service
metadata:
name: sflow-rt-sflow
spec:
type: NodePort
selector:
name: sflow-rt
ports:
- protocol: UDP
port: 6343
---
apiVersion: v1
kind: Service
metadata:
name: sflow-rt-rest
spec:
type: LoadBalancer
selector:
name: sflow-rt
ports:
- protocol: TCP
port: 8008
---
apiVersion: apps/v1
kind: Deployment
metadata:
name: sflow-rt
spec:
replicas: 1
selector:
matchLabels:
name: sflow-rt
template:
metadata:
labels:
name: sflow-rt
spec:
containers:
- name: sflow-rt
image: sflow/prometheus:latest
ports:
- name: http
protocol: TCP
containerPort: 8008
- name: sflow
protocol: UDP
containerPort: 6343
Run the Continue reading

SFMIX San Francisco shelter in place

A shelter in place order restricted San Francisco residents to their homes beginning at 12:01 a.m. on March 17, 2020. Many residents work for Bay Area technology companies such as Salesforce, Facebook, Twitter, Google, Netflix and Apple. Employees from these companies are able to, and have been instructed to, work remotely from their homes. In addition, other housebound residents are making use of social networking to keep in touch with friends and family as well as streaming media and online gaming for entertainment.

The traffic trend chart above from the San Francisco Metropolitan Internet Exchange (SFMIX) shows the change in network traffic that has resulted from the shelter in place order. Peak traffic has increased by around 10Gbit/s (a 25% increase) and continues throughout the day (whereas peaks previously occurred in the evenings).

The SFMIX network directly connects a number of data centers in the Bay Area and the member organizations that peer from those data centers.  Peering through the exchange network keeps traffic local by directly connecting companies with their employees and customers and avoiding potentially congested service provider networks.
SFMIX recently finished a network upgrade to 100Gbit/s Arista switches and all fiber Continue reading

Ubuntu 18.04

Ubuntu 18.04 comes with Linux kernel version 4.15. This version of the kernel includes efficient in-kernel packet sampling that can be used to provide network visibility for production servers running network heavy workloads, see Berkeley Packet Filter (BPF).
This article provides instructions for installing and configuring the open source Host sFlow agent to remotely monitor servers using the industry standard sFlow protocol. The sFlow-RT real-time analyzer is used to demonstrate the capabilities of sFlow telemetry.

Find the latest Host sFlow version on the Host sFlow download page.
wget https://github.com/sflow/host-sflow/releases/download/v2.0.25-3/hsflowd-ubuntu18_2.0.25-3_amd64.deb
sudo dpkg -i hsflowd-ubuntu18_2.0.25-3_amd64.deb
sudo systemctl enable hsflowd
The above commands download and install the software.
sflow {
collector { ip=10.0.0.30 }
pcap { speed=1G-1T }
tcp { }
systemd { }
}
Edit the /etc/hsflowd.conf file. The above example sends sFlow to a collector at 10.0.0.30, enables packet sampling on all network adapters, adds TCP performance information, and exports metrics for Linux services. See Configuring Host sFlow for Linux for the complete set of configuration options.
sudo systemctl restart hsflowd
Restart the Host sFlow daemon to start streaming telemetry to Continue reading

Docker testbed

The sFlow-RT real-time analytics platform receives a continuous telemetry stream from sFlow Agents embedded in network devices, hosts and applications and converts the raw measurements into actionable metrics, accessible through open APIs, see Writing Applications.

Application development is greatly simplified if you can emulate the infrastructure you want to monitor on your development machine. Mininet flow analyticsMininet dashboard, and Mininet weathermap describe how to use the open source Mininet network emulator to simulate networks and generate a live stream of standard sFlow telemetry data.

This article describes how to use Docker containers as a development platform. Docker Desktop provides a convenient method of running Docker on Mac and Windows desktops. These instructions assume you have already installed Docker.

First, find your host address (e.g. hostname -I, ifconfig en0, etc. depending on operating system), then open a terminal window and set the shell variable MY_IP:
MY_IP=10.0.0.70
Start a Host sFlow agent using the pre-built sflow/host-sflow image:
docker run --rm -d -e "COLLECTOR=$MY_IP" -e "SAMPLING=10" \
--net=host -v /var/run/docker.sock:/var/run/docker.sock:ro \
--name=host-sflow sflow/host-sflow
Note: Host, Docker, Swarm and Kubernetes monitoring describes how to deploy Host sFlow agents to monitor large scale Continue reading

CentOS 8

CentOS 8 / RHEL 8 come with Linux kernel version 4.18. This version of the kernel includes efficient in-kernel packet sampling that can be used to provide network visibility for production servers running network heavy workloads, see Berkeley Packet Filter (BPF).
This article provides instructions for installing and configuring the open source Host sFlow agent to remotely monitor servers using the industry standard sFlow protocol. The sFlow-RT real-time analyzer is used to demonstrate the capabilities of sFlow telemetry.

Find the latest Host sFlow version on the Host sFlow download page.
wget https://github.com/sflow/host-sflow/releases/download/v2.0.26-3/hsflowd-centos8-2.0.26-3.x86_64.rpm
sudo rpm -i hsflowd-centos8-2.0.26-3.x86_64.rpm
sudo systemctl enable hsflowd
The above commands download and install the software.
sflow {
collector { ip=10.0.0.30 }
pcap { speed=1G-1T }
tcp { }
systemd { }
}
Edit the /etc/hsflowd.conf file. The above example sends sFlow to a collector at 10.0.0.30, enables packet sampling on all network adapters, adds TCP performance information, and exports metrics for Linux services. See Configuring Host sFlow for Linux for the complete set of configuration options.
sudo systemctl restart hsflowd
Restart the Host sFlow daemon to Continue reading

SONiC

SONiC is part of the Open Compute Project (OCP), creating "an open source network operating system based on Linux that runs on switches from multiple vendors and ASICs." The latest SONiC.201911 release of the open source SONiC network operating system adds sFlow support.
SONiC: sFlow High Level Design
The diagram shows the elements of the implementation.
  1. The open source Host sFlow agent running in the sFlow container monitors the Redis database (in the Database container) for sFlow related configuration changes.
  2. The syncd container monitors the configuration database and pushes hardware settings (packet sampling) to the ASIC using the SAI (Switch Abstraction Inteface) driver (see SAI 1.5).
  3. The ASIC driver hands sampled packet headers and associated metadata captured by the ASIC to user space via the Linux PSAMPLE netlink channel (see Linux 4.11 kernel extends packet sampling support).
  4. The Host sFlow agent receives the PSAMPLE messages and forwards them to configured sFlow collector(s) as standard sFlow packet samples.
  5. In addition, the Host sFlow agent streams telemetry (interface counters and host metrics gathered from the Redis database and Linux kernel) to the collector(s) as standard sFlow counter records.
The following CLI commands enable sFlow Continue reading

Real-time DDoS mitigation using BGP RTBH and FlowSpec

DDoS Protect is a recently released open source application running on the sFlow-RT real-time analytics engine. The software uses streaming analytics to rapidly detect and characterize DDoS flood attacks and automatically applies BGP remote triggered black hole (RTBH) and/or FlowSpec controls to mitigate their impact. The total time to detect and mitigate an attack is in the order of a second.

The combination of multi-vendor standard telemetry (sFlow) and control (BGP FlowSpec) provide the real-time visibility and control needed to quickly and automatically adapt the network to address a range of challenging problems, including: DDoS, traffic engineering, and security.

Solutions are deployable today: Arista BGP FlowSpec describes the recent addition of BGP FlowSpec support to Arista EOS (EOS has long supported sFlow), and sFlow available on Juniper MX series routers describes the release of sFlow support on Juniper MX routers (which have long had BGP FlowSpec support). This article demonstrates DDoS mitigation using Arista EOS. Similar configurations should work with any router that supports sFlow and BGP FlowSpec.
The diagram shows a typical deployment scenario in which an instance of sFlow-RT (running the DDoS Protect application) receives sFlow from the site router (ce-router). A  Continue reading

SAI 1.5

The Open Compute Project (OCP), "is a rapidly growing community of engineers around the world whose mission is to design and enable the delivery of the most efficient server, storage and data center hardware designs available for scalable computing."

The OCP SAI (Switch Abstraction Interface) Project is an important part of the networking effort, defining "a vendor-independent way of controlling forwarding elements, such as a switching ASIC, an NPU or a software switch in a uniform manner." SAI 1.5 Release Notes describe enhancements to existing sFlow API, in particular adding support for the Linux psample netlink channel, see  Linux 4.11 kernel extends packet sampling support. Supporting the standard Linux interface for packet sampling simplifies the implementation of sFlow agents (e.g. Host sFlow) and ensures consistent behavior across hardware platforms to deliver real-time network-wide visibility using industry standard sFlow protocol.

Real-time monitoring at terabit speeds

The Flow Trend chart above shows a real-time, up to the second, view of nearly 3 terabits per second of traffic flowing across the SCinet network, described as the fastest, most powerful volunteer-built network in the world. The network is build each year to support The International Conference for High Performance Computing, Networking, Storage, and Analysis. The SC19 conference is currently underway in Denver, Colorado.
The diagram shows the Joint Big Data Testbed generating the traffic in the chart. The Caltech demonstration is described in NRE-19: SC19 Network Research Exhibition: Caltech Booth 543 Demonstrations Hosting NRE-13, NRE-19, NRE-20, NRE-22, NRE-23, NRE-24, NRE-35:
400GE First Data Networks: Caltech, Starlight/NRL, USC, SCinet/XNET, Ciena, Mellanox, Arista, Dell, 2CRSI, Echostreams, DDN and Pavilion Data, as well as other supporting optical, switch and server vendor partners will demonstrate the first fully functional 3 X400GE local ring network as well as 400GE wide area network ring, linking the Starlight and Caltech booths and Starlight in Chicago. This network will integrate storage using NVMe over Fabric, the latest high throughput methods, in-depth monitoring and realtime flow steering. As part of these demonstrations, we will make use of the latest DWDM, Waveserver Ai, and 400GE as Continue reading

SC19 SCinet: Grafana network traffic dashboard

The Grafana sFlow-RT Countries and Networks dashboard above shows traffic on the SCinet network, described as the fastest, most powerful volunteer-built network in the world. The network is build each year to support The International Conference for High Performance Computing, Networking, Storage, and Analysis. The SC19 conference is currently underway in Denver, Colorado and the screen capture is live data from the conference network.
The high speed switches and routers used to construct the SCinet network support industry standard sFlow streaming telemetry. In this case an instance of the sFlow-RT analytics engine receives the telemetry stream and generates flow analytics that are scraped every 15 seconds by an instance of the Prometheus time series database. The Prometheus database is in turn queried by an instance of Grafana which generated the dashboard shown at the top of the page.
In addition, sFlow-RT is running an embedded application that generates a real-time, up to the second, view of the traffic over the last 5 minutes.
This solution is extremely scalable. A single sFlow-RT instance, allocated only 1G of memory, easily monitors 158 network devices, while supporting 11 different applications (including the real-time dashboard and Prometheus export applications shown above).

Observability in Data Center Networks


Observability in Data Center Networks: In this session, you’ll learn how the sFlow protocol provides broad visibility in modern data center environments as they migrate to highly meshed topologies. Our data center workloads are shifting to take advantage of higher speeds and bandwidth, so visibility to east-west traffic within the data center is becoming more important. Join Peter Phaal—one of the inventors of sFlow—and Joe Reves from SolarWinds product management as they discuss how sFlow differs from other flow instrumentation to deliver visibility in the switching fabric.
THWACKcamp is SolarWinds’ free, annual, worldwide virtual IT learning event connecting thousands of skilled IT professionals with industry experts and SolarWinds technical staff. This video was one of the sessions.

InfluxDB 2.0

Introducing the Next-Generation InfluxDB 2.0 Platform mentions that InfluxDB 2.0 will be able to scrape Prometheus exporters. Get started with InfluxDB provides instructions for running an alpha version of the new software using Docker:
docker run --name influxdb -p 9999:9999 quay.io/influxdb/influxdb:2.0.0-alpha
Prometheus exporter describes an application that runs on the sFlow-RT analytics platform that converts real-time streaming telemetry from industry standard sFlow agents. Host, Docker, Swarm and Kubernetes monitoring describes how to deploy agents on popular container orchestration platforms.
The screen capture above shows three scrapers configured in InfluxDB 2.0:
  1. sflow-rt-analyzer,
    URL: http://10.0.0.70:8008/prometheus/analyzer/txt
  2. sflow-rt-dump,
    URL: http://10.0.0.70:8008/prometheus/metrics/ALL/ALL/txt
  3. sflow-rt-flow-src-dst,
    URL: http://10.0.0.70:8008/app/prometheus/scripts/export.js/flows/ALL/txt?metric=flow_src_dst_bps&key=ipsource,ipdestination&value=bytes&aggMode=max&maxFlows=100&minValue=1000&scale=8
The first collects metrics about the performance of the sFlow-RT analytics engine, the second, all the metrics exported by the sFlow agents, and the third, is a flow metric (see Flow metrics with Prometheus and Grafana).

Updated 19 October 2019, native support for Prometheus export added to sFlow-RT, URLs 1 and 2 modified to reflect new API.
InfluxDB 2.0 now includes the data exploration and dashboard building capabilities that were previously in the separate Chronograf application. The screen Continue reading

Flow metrics with Prometheus and Grafana

The Grafana dashboard above shows real-time network traffic flow metrics. This article describes how to define and collect flow metrics using the Prometheus time series database and build Grafana dashboards using those metrics.
Prometheus exporter describes an application that runs on the sFlow-RT analytics platform that converts real-time streaming telemetry from industry standard sFlow agents. Host, Docker, Swarm and Kubernetes monitoring describes how to deploy agents on popular container orchestration platforms.

The latest version of the Prometheus exporter application adds flow export.
global:
scrape_interval: 15s
evaluation_interval: 15s

rule_files:
# - "first.rules"
# - "second.rules"

scrape_configs:
- job_name: 'sflow-rt-metrics'
metrics_path: /prometheus/metrics/ALL/ALL/txt
static_configs:
- targets: ['10.0.0.70:8008']
- job_name: 'sflow-rt-src-dst-bps'
metrics_path: /app/prometheus/scripts/export.js/flows/ALL/txt
static_configs:
- targets: ['10.0.0.70:8008']
params:
metric: ['ip_src_dst_bps']
key: ['ipsource','ipdestination']
label: ['src','dst']
value: ['bytes']
scale: ['8']
minValue: ['1000']
maxFlows: ['100']
- job_name: 'sflow-rt-countries-bps'
metrics_path: /app/prometheus/scripts/export.js/flows/ALL/txt
static_configs:
- targets: ['10.0.0.70:8008']
params:
metric: ['ip_countries_bps']
key: ['null:[country:ipsource]:unknown','null:[country:ipdestination]:unknown']
label: ['src','dst']
value: ['bytes']
scale: ['8']
aggMode: ['sum']
minValue: ['1000']
maxFlows: ['100']
The above prometheus.yml file extends the previous example to add two additional scrape jobs, sflow-rt-src-dst-bps and sflow-rt-countries-bps, that return flow metrics. Defining flows describes the attributes and settings available to build Continue reading

Host, Docker, Swarm and Kubernetes monitoring

The open source Host sFlow agent incorporates technologies that address the challenges of microservice monitoring; leveraging recent enhancements to Berkeley Packet Filter (BPF) in the Linux kernel to randomly sample packets, and  Asynchronous Docker metrics to track rapidly changing workloads. The continuous stream of real-time telemetry from all compute nodes, transported using the industry standard sFlow protocol, provides comprehensive real-time cluster-wide visibility into all services and the traffic flowing between them.

The Host sFlow agent is available as pre-packaged rpm/deb files that can be downloaded and installed on each node in a cluster.
sflow {
collector { ip=10.0.0.70 }
docker { }
pcap { dev=docker0 }
pcap { dev=docker_gwbridge }
}
The above /etc/hsflowd.conf file, see Configuring Host sFlow for Linux via /etc/hsflowd.conf, enables the docker {} and pcap {} modules for detailed visibility into container metrics and network traffic flows, and streams telemetry to an sFlow collector (10.0.0.70). The configuration is the same for every node making it simple to install and configure Host sFlow on all nodes using orchestration software such as Puppet, Chef, Ansible, etc.

The agent is also available as the pre-build sflow/host-sflow image, Continue reading

Packet analysis using Docker

Why use sFlow for packet analysis? To rephrase the Heineken slogan, sFlow reaches the parts of the network that other technologies cannot reach. Industry standard sFlow is widely supported by switch vendors, embedding wire-speed packet monitoring throughout the network. With sFlow, any link or group of links can be remotely monitored. The alternative approach of physically attaching a probe to a SPAN/Mirror port is becoming much less feasible with increasing network sizes (10's of thousands of switch ports) and link speeds (10, 100, and 400 Gigabits). Using sFlow for packet capture doesn't replace traditional packet analysis, instead sFlow extends the capabilities of existing packet capture tools into the high speed switched network.

This article describes the sflow/tcpdump  and sflow/tshark Docker images, which provide a convenient way to analyze packets captured using sFlow.

Run the following command to analyze packets using tcpdump:
$ docker run -p 6343:6343/udp -p 8008:8008 sflow/tcpdump

19:06:42.000000 ARP, Reply 10.0.0.254 is-at c0:ea:e4:89:b0:98 (oui Unknown), length 64
19:06:42.000000 IP 10.0.0.236.548 > 10.0.0.70.61719: Flags [P.], seq 3380015689:3380015713, ack 515038158, win 41992, options [nop,nop,TS val 1720029042 ecr 904769627], length 24
19:06:42.000000 Continue reading
1 2 3 10