Author Archives: Anton Karneliuk
Author Archives: Anton Karneliuk
Hello my friend,
Network Function Virtualisation (NFV) is not a new topic. There are numerous blogpost and articles, even in our blog, which review this topic. Yet, there is much more we can cover. Today we’ll share some insights on one of the very interesting products existing on the market today: 6WIND vRouter Turbo Router. We have a limited amount of days to write a few articles under our evaluation license. Hence, we’ll focus only on the most critical elements.
1
2
3
4
5 No part of this blogpost could be reproduced, stored in a
retrieval system, or transmitted in any form or by any
means, electronic, mechanical or photocopying, recording,
or otherwise, for commercial purposes without the
prior permission of the author.
It absolutely is. In fact, Linux is the real home for automation systems, as in many cases it hosts the tools you create in Ansible, Python, Bash, Go or any other language. At the same time, in order to effectively work with Linux, you need to know how to automate management and operation of Linux operating system itself. And you will be absolutely capable to do that, once you attend our Continue reading
Hello my friend,
Thanks a lot for all your interactions over our last post, it was very good feeling. As it attracted your attention, we decided to explore this topic further. Today we’ll focus on two main aspects: how you can figure out from Nokia SR OS CLI, what is your path to use in pySROS and how you visualise a configuration tree following YANG module.
1
2
3
4
5 No part of this blogpost could be reproduced, stored in a
retrieval system, or transmitted in any form or by any
means, electronic, mechanical or photocopying, recording,
or otherwise, for commercial purposes without the
prior permission of the author.
Facebook outage happened the last week caused a lot of discussions, how reliable automation is in general and if there are too much automatons these days already in the infrastructure systems. Really, if Facebook with their smartest network and automation engineers failed, should I stay far away from automation? We believe, everyone will find his or her answer. From our perspective, we know that Facebook will make their systems after failure more robust and working. By the way, in our training Continue reading
Hello my friend,
For quite a while we were silent (literally, the last post was month ago), as we were busy with extremely busy with completion of an important customer engagement. It was successfully completed and we are back on track with writing new interesting and useful (we hope so :-)) materials for you! And today we talk about our favourite topic – automation. To be precise, automation of Nokia SR OS devices with a new Python library called pySROS created by Nokia folks.
1
2
3
4
5 No part of this blogpost could be reproduced, stored in a
retrieval system, or transmitted in any form or by any
means, electronic, mechanical or photocopying, recording,
or otherwise, for commercial purposes without the
prior permission of the author.
If you have ever been in a situation, when there are much more work than hours in a day, then you know how important is to be able to do the job quickly and accurately. It is even better, if such job is done not by yourself, but rather by someone else or something else. This is exactly, where the automation of network and Continue reading
Hello my friend,
It is been a while since we posted our last blogpost, which was touching Infrastructure aspects of building Multi Server Cloud with ProxMox. After summer break we continue our blogging and developing activities at Karneliuk.com. Today we’ll show some backstage of the software development of the pygnmi.
1
2
3
4
5 No part of this blogpost could be reproduced, stored in a
retrieval system, or transmitted in any form or by any
means, electronic, mechanical or photocopying, recording,
or otherwise, for commercial purposes without the
prior permission of the author.
In a nutshell, Automation and Network Automation are just subset of tasks from a broader topic called Software Development. With the move of the world towards Industry 4.0, the digital economy growth and others, all sort of applications are becoming more widely used. In order you create a history, and not being a part of that in the past, you shall know the principles of software development and be able to create applications yourselves. Sounds complicated? It may be, indeed. However, with our Network Automation Training you definitely will have a break-through in the software development world.
At our Continue reading
Hello my friend,
In the previous blogpost we covered the installation of Proxmox as a core platform for building open source virtualisation environment. Today we’ll continue this discussion and will show how to create a multi server cloud in order to better spread the load and provide resiliency for your applications.
1
2
3
4
5 No part of this blogpost could be reproduced, stored in a
retrieval system, or transmitted in any form or by any
means, electronic, mechanical or photocopying, recording,
or otherwise, for commercial purposes without the
prior permission of the author.
In many cases, Linux is a major driving power behind modern clouds. In fact, if you look across all current big clouds, such as Amazon Web Services, Google Cloud Platform, Microsoft Azure, you will see Linux everywhere: on servers and on network devices (e.g., data centre switches). Therefore, knowledge how to deal with Linux and how to automate it is crucial to be successful in automation current IT systems.
At our trainings, advanced network automation and automation with Nornir (2nd step after advanced network automation), we give you detailed knowledge of all the technologies relevant:
Hello my friend,
Just the last week we finished our Zero-to-Hero Network Automation Training, which was very intensive and very interesting. The one could think: it is time for vacation now!.. Not quite yet. We decided to use the time wisely and upgrade our lab to bring possibilities for customers to use it. Lab upgrade means a major infrastructure project, which involves brining new hardware, changing topology and new software to simplify its management. Sounds interesting? Jump to details!
1
2
3
4
5 No part of this blogpost could be reproduced, stored in a
retrieval system, or transmitted in any form or by any
means, electronic, mechanical or photocopying, recording,
or otherwise, for commercial purposes without the
prior permission of the author.
Each and every element of your entire IT landscape requires two actions. It shall be monitored and it shall be managed. Being managed means that the element shall be configured and this is the first step for all sort of automations. Configuration management is a perfect use case to start automating your infrastructure, which spans servers, network devices, VMs, containers and much more. And we are here to help you to do Continue reading
Hello my friend,
Talking to some our partners, they mentioned that they are extensively using the unit tests python libraries for the development of the software in Python. They are creating their applications in Django framework; therefore, they are using a built-in Django testing library, which is based on Python’s standard unittest. We start looking, how we can use that ourselves…
1
2
3
4
5 No part of this blogpost could be reproduced, stored in a
retrieval system, or transmitted in any form or by any
means, electronic, mechanical or photocopying, recording,
or otherwise, for commercial purposes without the
prior permission of the author.
What do we, network engineers, do to quickly check that our network devices are reachable? We do ping. However, is that really good enough to check that your devices are OK? The one could say: I don’t even use a ping for that, as I rely on my NMS. This is also a good answer. In the same time, though, the vast majority of the NMS doesn’t contain all the details of network devices configuration and their operational data: e.g. all the details of OSPF and Continue reading
Hello my friend,
Very often at our zero-to-hero network automation training we are asked, what is the benefit of Bash? Why do we need to know Bash, if everyone is talking about Bash? The question is absolutely legitimate. One of the interesting and also legitimate answers I’ve just figured the last week.
1
2
3
4
5 No part of this blogpost could be reproduced, stored in a
retrieval system, or transmitted in any form or by any
means, electronic, mechanical or photocopying, recording,
or otherwise, for commercial purposes without the
prior permission of the author.
Everything in your network and IT operational processes can be automated. It is just matter of the resources (time, money, efforts) you spent on that against the gain you are obtaining. Basically, that is something what financial guys and girls call ROI (Return On Investments). The bigger the outcome and the lower the effort, the better the overall automation solution.
In our automation trainings we explain various use cases and success (and failure) strategies, how to build the (network) automation systems and tools, to make sure your gain is maximum. Our instructors have an extensive experience building Continue reading
Hello my friend,
In the previous blogpost we’ve introduced the Batfish and how to set it up. Today we’ll take a look how to perform the analysis of the configuration to figure out discrepancies, which may lead to broken operation of your network.
1
2
3
4
5 No part of this blogpost could be reproduced, stored in a
retrieval system, or transmitted in any form or by any
means, electronic, mechanical or photocopying, recording,
or otherwise, for commercial purposes without the
prior permission of the author.
In software development we have a concept called CI/CD (Continuous Integration/Continuous Delivery). In a nutshell, it’s a methodology, which incorporates mandatory testing of configuration (code, software version, etc) before bringing it to production. The main idea behind it is that automated testing and validation will make sure that code is stable and fit for purpose. Automated testing? That’s where the automation comes to the stage.
And automation is something what we are experts in. And you can benefit from that expertise as well.
In our network automation training we follow zero to hero approach, where we start with the basics including Linux operation and administration topped with Continue reading
Hello my friend,
The new year we start with a new topic, which is a configuration analysis of the multivendor networks. We have a passion both to create our own open source tools and to use existing, creating by other teams and project. Today we will start dive in one of such a tool.
1
2
3
4
5 No part of this blogpost could be reproduced, stored in a
retrieval system, or transmitted in any form or by any
means, electronic, mechanical or photocopying, recording,
or otherwise, for commercial purposes without the
prior permission of the author.
In software development we have a concept called CI/CD (Continuous Integration/Continuous Delivery). In a nutshell, it’s a methodology, which incorporates mandatory testing of configuration (code, software version, etc) before bringing it to production. The main idea behind it is that automated testing and validation will make sure that code is stable and fit for purpose. Automated testing? That’s where the automation comes to the stage.
And automation is something what we are experts in. And you can benefit from that expertise as well.
In our network automation training we follow zero to hero approach, where we Continue reading
Hello my friend,
Have we thought 5 years ago that we would manage to keep the blog running for such a long period of time and would create such a huge amount of useful information for the community? Definitely not. And that’s why it is even important we managed to achieve such a milestone. Thank a lot each of you for your support, ideas, feedbacks, shared on social medias and likes! That means a lot for us. Please, continue doing so
The last 12 month were absolutely incredible. The COVID-19 pandemic struck word so much that noone could ever believe. As a result, we, pretty much as a whole work, worked from home all past 12 moths… Which was an unusual experience. However, we used this time (in fact, due to no necessity to spend time on commute, we have more time to work) productively working on various projects for you, our dear readers. So, what have we managed to achieve?
One of the massive achievements was to complete the book Network Automation and Programmability Fundamentals, which was published by Cisco Press in May 2021. The journey for us started back in Continue reading
Hello my friend,
Over the past several articles we have taken a look into the most popular tools for the network troubleshooting. To be precise, we have covered an approach to find your public IP address in a programmable way, how to measure your Internet connectivity speed with Speedtest, how to measure the connectivity speed between your endpoints using iPerf, rapid check of hosts presence in the network with fping, how to check the quality of the path through the network with MTR, and how to see what are in the packets you have in your network. Covering these tools for you and sharing ideas about automation with them led us to creating a new open source tool, which we are happy to share with the community. This time we decided not to think long on its name and just called it the Traceroute Analyzer.
1
2
3
4
5 No part of this blogpost could be reproduced, stored in a
retrieval system, or transmitted in any form or by any
means, electronic, mechanical or photocopying, recording,
or otherwise, for commercial purposes without the
prior permission of the author.
Hello my friend,
When something goes wrong with the distributed application, where the network is involved (e.g., between client and web service, or between frontend and backend of services), the network is a first thing to be blamed. After the troubleshooting, it is often turned out that the network is innocent, but we need first need to prove it.
1
2
3
4
5 No part of this blogpost could be reproduced, stored in a
retrieval system, or transmitted in any form or by any
means, electronic, mechanical or photocopying, recording,
or otherwise, for commercial purposes without the
prior permission of the author.
The truth is that automation helped me so many times to figure out the root cause of the network outages or malfunctions that I even stopped counting that. I may say that automaton solutions work perfect, if you create them to solve your issues and tailor to your environment.
That’s what our Live Network Automation Training (10 weeks) and Automation with Nornir (2 weeks) are all about: to show you real automation in a real environment with multiple vendors together. No matter what those vendors are, the automation principles, tools Continue reading
Hello my friend,
For a long time we haven’t posted blogs about pure network technologies. However, recently we were working on some interesting use case, which so far is not yet covered at a level of the working details nowhere in the internet. As such, we decided to share with you our findings and working details.
1
2
3
4
5 No part of this blogpost could be reproduced, stored in a
retrieval system, or transmitted in any form or by any
means, electronic, mechanical or photocopying, recording,
or otherwise, for commercial purposes without the
prior permission of the author.
In some (big) networks the BGP-SR-TE is a good signal the SR-TE policies, so that the PE routers can build the SR-TE tunnels without the need to configure them locally. However, the BGP-SR-TE requires a full pledged SDN controller, so that you can generate the SR-TE policy in the backend using some UI/API and send the policies down to the network elements using the BGP. If you want to have a somewhat simpler setup, you may need need to deploy the tunnels manually. In this case, the automation is your closest Continue reading
Hello my friend,
Some time ago in LinkedIn we announced that we are working on the tool, which will allows you to model and analyse your network. As one of our primary focuses is data centres, we started from there. Despite it is an early stages, but we are happy and proud to introduce you HAWK: Highly-efficient Automated Weapon Kit. For now, this is a collection of the tools for the network management and analysis, but probably later we will put it under a joint hoot of some front-end, who knows…
1
2
3
4
5 No part of this blogpost could be reproduced, stored in a
retrieval system, or transmitted in any form or by any
means, electronic, mechanical or photocopying, recording,
or otherwise, for commercial purposes without the
prior permission of the author.
In order to automate any network operation, you need to write a script, even if that is a simple one. On the other hand, any script is a program or software. This means that the creating of the scripts for the network automation is a form of the software development. And it is fun. Continue reading
Hello my friend,
after show pause we continue our blog series about the most popular network troubleshooting tools, which humankind has ever created. Today we’ll take a look at one of the most useful tool to obtain the information about the path between two endpoints and possible packet drops over that path. Ladies and gentlemen, please, welcome MTR.
1
2
3
4
5 No part of this blogpost could be reproduced, stored in a
retrieval system, or transmitted in any form or by any
means, electronic, mechanical or photocopying, recording,
or otherwise, for commercial purposes without the
prior permission of the author.
In case of the infrastructure problems (networks, servers, VMs, containers), the time matters a lot. The quicker we can find the issue and fix that, the better it will be for our applications and our customers. Automation without doubts one of the key components, which allows you to quickly find and fix your issues.
In our trainings, the Live Network Automation Training (10 weeks) and Automation with Nornir (2 weeks), we explore a lot of real use cases, where the automation helps you to validate the state of you Continue reading
Hello my friend,
Recently we were asked, what is the safest way to store the credentials for network devices to your automation tools (e.g., the one based on Python and gNMI). Building the network automation solutions for a while, we have a good answer to you.
1
2
3
4
5 No part of this blogpost could be reproduced, stored in a
retrieval system, or transmitted in any form or by any
means, electronic, mechanical or photocopying, recording,
or otherwise, for commercial purposes without the
prior permission of the author.
gNMI was created by Google to manage their data centres and backbone network and is widely used by other biggest companies worldwide. However, it doesn’t mean that only the big guys can benefit from that. Every company and network can get the advantage of a single protocol for the configuration, operation, and streaming telemetry in their network provided your network devices support that.
At our trainings, advanced network automation and automation with Nornir (2nd step after advanced network automation), we give you detailed knowledge of all the technologies relevant:
Hello my friend,
Continuing our explanation of the pyGNMI, we’ll take a loon into the security aspect of the tool. Namely, we will take a look how quickly and easily you can implement the encryption between your host running pyGNMI and the gNMI speaking network function.
1
2
3
4
5 No part of this blogpost could be reproduced, stored in a
retrieval system, or transmitted in any form or by any
means, electronic, mechanical or photocopying, recording,
or otherwise, for commercial purposes without the
prior permission of the author.
The automation world (network and not only) can at a high level be split into text-based automation and model-driven automation. The text based automation is all about Linux systems, where we typically template the whole configuration files, put them in the corresponding folders and then restart daemons. The model driven approach is where we communicate with the network devices using the NETCONF, RESTCONF or gNMI based on the YANG modules. At the current moment, gNMI is most dynamically developing protocol. We not only teaches you how it works and when to use it, but we also created a Python library, Continue reading
Hello my friend,
In one of the previous blogposts we have share the details how you can build the containerised PKI relying Docker, Alpine Linux and OpenSSL. Today we’ll show how you can use it.
1
2
3
4
5 No part of this blogpost could be reproduced, stored in a
retrieval system, or transmitted in any form or by any
means, electronic, mechanical or photocopying, recording,
or otherwise, for commercial purposes without the
prior permission of the author.
In one of the recent articles dedicated to threats to the national security with respect to 5G caused by the current operation models in the big service providers, one the network automation were named as one of the key mitigation approaches. The reason is very simple: network automation allows you to keep (or even increase) the pace of the network changes with increasing the stability without increasing the size of your network operation teams.
We absolutely agree with this statement. We have observed ourselves that in many cases the key technical competence for service providers is being outsourced and in certain cases the service providers have to rely on the external parties to perform Continue reading
Hello my friend,
Quite often, when we do the troubleshooting of our networks and systems, we want to figure out, which hosts are alive in the certain range. The quickest and the easiest way (though, not 100% accurate) is to run the ping against a specific range of IPs. There is a brilliant tool for this purpose, which is called fping.
1
2
3
4
5 No part of this blogpost could be reproduced, stored in a
retrieval system, or transmitted in any form or by any
means, electronic, mechanical or photocopying, recording,
or otherwise, for commercial purposes without the
prior permission of the author.
All the time, when the outage is happening in the production environment, either with the network or server/application infrastructure, the race starts to restore the service as soon as possible. Automation is a key helper there.
In our trainings, the Live Network Automation Training (10 weeks) and Automation with Nornir (2 weeks), we explore a lot of real use cases, where the automation helps you to validate the state of you network and change it if necessary. You will learn the whole spectre of Continue reading