Links in Virtual Labs

There are three major ways to connect network devices in the physical world:

  • Point-to-point links between devices (usually using some variant of Ethernet)
  • Multi-access layer-1 networks running some IEEE 802.x encapsulation on top of that (GPON, WiFi, Ethernet hubs)
  • Multi-access switched layer-2 network (dumb switches, hopefully running some STP variant)

Implementing these connections in virtual labs is a bit harder than one might think, as all virtualization solutions assume you plan to run virtual servers connected to Ethernet segments.

Read more …

Setting up Proxmox Backup Server

Setting up Proxmox Backup Server

If you follow my blog, you may know that I recently migrated to Proxmox as my hypervisor, and I absolutely love it. My Proxmox server runs on a custom-built PC with an i9-10900K, 128GB RAM, and 3 x 1TB SSDs. I’m not using anything fancy, just a good number of VMs spread across two of the SSDs.

In this post, we’ll go through the process of setting up Proxmox Backup Server and backing up all the VMs from my Proxmox server to this backup server.

As always, if you find this post helpful, press the ‘clap’ button. It means a lot to me and helps me know you enjoy this type of content.

Running EVE-NG in Proxmox
In this blog post, we’ll go through the steps to install EVE-NG in Proxmox. Most of the VM’s settings can be left at their default values, but there are a couple of changes I had to make.
Setting up Proxmox Backup ServerPacketswitchSuresh Vina
Setting up Proxmox Backup Server

Why Proxmox Backup Server?

I’ve been running Proxmox for a few months without any backups. I did create some snapshots inside Proxmox in case I messed up a VM, so I could roll back if needed. But this was Continue reading

From Python to Go 013. Handling Errors And Exceptions.

Hello my friend,

One of the important aspects of writing any application is to ensure that it can operate, as long it is safe, when “something” went wrong during its execution. This “something” can really be anything: it can be wrong user input, it could be temporary unreachable network device or service or API endpoint, it could be missing file or incorrect path. In certain circumstance, for example when we talk about API gateways and web servers, it becomes even more critical. In today’s blog post we’ll see how to handle errors/exceptions in Python and Go (Golang).

How Automation Is Important?

I recently worked on a big project, which involves a lot of moving parts. To make these parts moving smoothly, I needed to analyze and compare data across multiple systems before making a decision. The amount of data is huge, thousands of line of data in every system and it is very easy to make mistake, which will impact users. How can I be sure I don’t miss anything? Scripting and automating! I’ve developed a tool, which requests via APIs data from multiple sources, analyses it based on my criteria and collates final report.

And you can build such Continue reading

AI Metrics

AI Metrics is available on GitHub. The application provides performance metrics for AI/ML RoCEv2 network traffic, for example, large scale CUDA compute tasks using NVIDIA Collective Communication Library (NCCL) operations for inter-GPU communications: AllReduce, Broadcast, Reduce, AllGather, and ReduceScatter.

The dashboard shown above is from a simulated network 1,000 switches, each with 48 ports access ports connected to a host. Activity occurs in a 256mS on / off cycle to emulate an AI learning run. The metrics include:

  • Total Traffic Total traffic entering fabric
  • Operations Total RoCEv2 operations broken out by type
  • Core Link Traffic Histogram of load on fabric links
  • Edge Link Traffic Histogram of load on access ports
  • RDMA Operations Total RDMA operations
  • RDMA Bytes Average RDMA operation size
  • Credits Average number of credits in RoCEv2 acknowledgements
  • Period Detected period of compute / exchange activity on fabric (in this case just over 0.5 seconds)
  • Congestion Total ECN / CNP congestion messages
  • Errors Total ingress / egress errors
  • Discards Total ingress / egress discards
  • Drop Reasons Packet drop reasons

Note: Clicking on peaks in the charts shows values at that time.

This article gives step-by-step instructions to run the AI Metrics application in a production environment and integrate Continue reading

Running Unifi Network Application in Docker

Running Unifi Network Application in Docker

As of January 2024, the linuxserver/unifi-controller Docker image has been deprecated and is no longer receiving updates. This means it may no longer be secure or compatible with newer versions of UniFi devices. Users are now advised to switch to the linuxserver/unifi-network-application image, which is actively maintained and provides the latest version of the UniFi Network Application.

In this blog post, we will cover how to run the new image in Docker and how to migrate from the old image to the new one without losing your existing configuration.

As always, if you find this post helpful, press the ‘clap’ button on the left. It means a lot to me and helps me know you enjoy this type of content.

Overview

In my homelab, I have a USW Lite 8 PoE, U6-Pro, and U6 Mesh. I had been running the older linuxserver/unifi-controller Docker image since I first set them up. My controller was on version 8.0.34, if I remember correctly.

Recently, I moved to a new host, so I decided it was the right time to migrate to the new linuxserver/unifi-network-application image. While checking their GitHub page, I found this note.

You Continue reading

TNO014: Beneath the Cable Color – A Look at The Siemon Company’s High Speed Cable Assemblies (Sponsored)

Today we get an education on what a high speed cable assembly really is from guest Ryan Harris from today’s sponsor, The Siemon Company.  Ryan shares the evolution of high-speed pluggable cable assemblies from 64kbps ISDN B channels to 800 gigabit connections, and describes how they work. We discuss the Siemon Company’s history and its... Read more »

Intel Pushes Out “Clearwater Forest” Xeon 7, Sidelines “Falcon Shores” Accelerator

If Intel hopes to survive the next few years as a freestanding company and return to its role as innovator, it can not afford to waste its time and it cannot afford to make any more mistakes. …

Intel Pushes Out “Clearwater Forest” Xeon 7, Sidelines “Falcon Shores” Accelerator was written by Timothy Prickett Morgan at The Next Platform.

Hedge 257: Gaining Experience in IT

Every recruiter and hiring manager wants people with five years of experience, but you cannot get experience without being hired into a position. How can you break this conundrum? Daniel Dib joins Tom and Russ to talk about how folks just coming into IT, or even those with lots of experience who are trying to shift their focus, can gain experience.

download

netlab 1.9.4: Bug fixes, VRRPv3 on Junos

During the last three weeks, we were busy squashing bugs (device configuration fixes, other bug fixes). Some were recent; others were ancient pests uncovered by better integration tests. The end result: netlab release 1.9.4.

netlab release 1.9.4 passed hundreds of integration tests and should be a better choice than the previous 1.9 releases. To upgrade, execute pip3 install --upgrade networklab.

New to netlab? Start with the Getting Started document and the installation guide, or run it in a GitHub codespace.

Update: 2025-02-03

We still missed a few quirks :( Release 1.9.4-post1 addresses those (and, unfortunately, I’m pretty sure there will be more).

Replay pcap files using sflowtool


It can be very useful to capture sFlow telemetry from production networks so that it can be replayed later to perform off-line analysis, or to develop or evaluate sFlow collection tools. 
sudo tcpdump -i any -s 0 -w sflow.pcap udp port 6343
Run the command above on the system you are using to collect sFlow data (if you aren't yet collecting sFlow, see Agents for suggested configuration settings). Type Control-C to end the capture after 5 to 10 minutes.  Copy the resulting sflow.pcap file to your laptop. 
docker run --rm -it -v $PWD/sflow.pcap:/sflow.pcap sflow/sflowtool \
  -r /sflow.pcap -P 1
Either compile the latest version of sflowtool or, as shown above, use Docker to run the pre-built sflow/sflowtool image. The -P (Playback) option replays the trace in real-time and displays the contents of each sFlow message. Running sflowtool using Docker provides additional examples, including converting the sFlow messages into JSON format for processing by a Python script. 
docker run --rm -it -v $PWD/sflow.pcap:/sflow.pcap sflow/sflowtool \
  -r /sflow.pcap -f 192.168.4.198/6343 -P 1
The -f (forwarding) option takes an IP address and UDP port number as arguments, in this Continue reading

Using the Continue VSCode Extension and Local LLMs for Improved Coding

Using the Continue VSCode Extension and Local LLMs for Improved Coding

Welcome back to another post on local LLMs. In this post, we’ll look at setting up a fully local coding assistant inside VSCode using the Continue extension and Ollama. Let’s get started.

As always, if you find this post helpful, press the ‘clap’ button on the left. It means a lot to me and helps me know you enjoy this type of content.

Overview

We’ve covered Ollama and Local LLMs in previous blog posts (linked below), but here’s a quick summary.

Ollama is a tool that lets you run large language models (LLMs) directly on your local machine. Local LLMs are language models that run on your computer instead of relying on cloud-based services like ChatGPT. This means you can use them without sending your data to external servers, which is great for privacy. They also work offline, so you’re not dependent on an Internet connection.

That said, it’s important to note that local models, especially on smaller setups, won’t match the speed or performance of cloud-based models like ChatGPT. These cloud models are powered by massive infrastructure, so they’re faster and often more accurate. However, the trade-off is privacy and offline access, which local LLMs provide.

In Continue reading

Using Ollama with a Web-Based GUI

Using Ollama with a Web-Based GUI

When I first started using local LLMs with Ollama, I quickly realised it relies on a command-line interface to interact with the models. It also comes with an API, but let’s be honest, most of us, myself included, prefer a GUI, much like the one ChatGPT provides. There are plenty of options available, but I decided to try Open Web GUI. In this blog post, we’ll explore what Open-WebGUI is and how simple it is to set up a web-based interface for your local LLMs.

As always, if you find this post helpful, press the ‘clap’ button on the left. It means a lot to me and helps me know you enjoy this type of content.

Overview

Ollama is a tool for running local LLMs, offering privacy and control over your data. Out of the box, it lets you interact with models via the terminal or through its API. Installing Ollama is straightforward, and if you’d like a detailed guide, check out my other blog post which is linked below.

This blog post assumes you already have Ollama set up and running. For reference, I’m running this on my MacBook (M3 Pro with 18GB of RAM).

open-webui

Open Continue reading

Running Large Language Models (LLM) on Your Own Machine Using Ollama

Running Large Language Models (LLM) on Your Own Machine Using Ollama

I’m going to start by saying I’m totally new to LLMs and running them locally, so I’m not going to pretend like I know what I am doing. I’ve been learning about Ollama for some time now and thought I would share it with my readers as always. This is such an interesting topic and I’m ready to go into the rabbit hole.

As always, if you find the content useful, don’t forget to press the ‘clap’ button to your left. This is one way for me to know that you like this type of content, which means a lot to me. So, let's get started.

Large Language Models (LLMs)

LLMs, or Large Language Models, are a type of artificial intelligence designed to process and generate natural language. They are trained on vast amounts of text data, enabling them to understand context, identify patterns, and produce human-like responses. These models can perform various tasks such as answering questions, translating languages, summarising text, generating creative content, and assisting with coding. LLMs have gained significant attention in recent years due to their impressive performance and versatility.

Palo Alto URL Filtering and SSL Decryption
Without SSL decryption, the Palo Alto firewall Continue reading

N4N011: What’s the Difference Between LAG, MLAG, MC-LAG, and Stacking?

In today’s episode, we address listener Kieren’s question about the differences between LAG, MLAG, MC-LAG, and stacking. We tackle the nuances of Link Aggregation (LAG) and the Link Aggregation Control Protocol (LACP), and explain their roles in redundancy and bandwidth efficiency. We also discuss the complexities and differences among vendors and overall benefits of Multi-Chassis... Read more »

IBM Takes The Patient Path To Future GenAI Profits

While the hyperscalers and cloud builders provide the best indicator of what it takes to create state of the art GenAI models and the infrastructure to train them as well as to put them into production for practical use through an API interface, perhaps IBM is one of the best leading indicators for how GenAI will slowly be adopted by the enterprises of the world within their own organizations. …

IBM Takes The Patient Path To Future GenAI Profits was written by Timothy Prickett Morgan at The Next Platform.

The Curious Case of the BGP Connect State

I got this question from Paul:

Have you ever seen a BGP peer in the “Connect” state? In 20 years, I have never been able to see or reproduce this state, nor any mention in a debug/log. I am starting to believe that all the documentation is BS, and this does not exist.

The BGP Finite State Machine (FSM) (at least the one defined in RFC 4271 and amended in RFC 9687) is “a bit” hard to grasp but the basics haven’t changed from the ancient days of RFC 1771:

Read more …
1 … 109 110 111 112 113 … 3,873