Hey Everyone,
Last week we covered all things KubeCon which has already kicked off with some co-located events in Amsterdam, I am not going to be heading out till really early tomorrow morning, just in time for our learning day event.
Kubernetes 1.27 Release
The first release of 2023! 60 enhancements. 18 of those enhancements are entering Alpha, 29 are graduating to Beta, and 13 are graduating to Stable.
The biggest aspect of this release was the replacement of the old image registry. The new registry registry.k8s.io they are asking that all users migrate from k8s.gcr.io as soon as possible.
I am going to focus on some of the key storage enhancements here.
ReadWriteOncePod PersistentVolume access
Back in 1.22, we got a new access mode “ReadWriteOncePod” which enables you to restrict volume access to a single pod in the cluster, ensuring that only one pod can write to the volume at a time. This can be particularly useful for stateful workloads that require single-writer access to storage.
This has now moved to the beta stage, which allows higher-priority pods to pre-empt lower-priority pods.
Robust VolumeManager reconstruction
Along with the ReadWriteOncePod enhancement, there is a new volume manager in town, well at the beta stage. Pre 1.25, kubelet used different default behaviour for discovering mounted volumes during start-up.
This new feature will provide us with populating additional information about how existing volumes are mounted during the kubelet start-up. This is going to help with volume clean up.
VolumeGroupSnapshot
This one is graduating to Alpha which adds support in the Kubernetes API to create a snapshot of multiple volumes together, in a way that they are consistent, and prevent data loss.
For those that have been in the virtualisation space for a while you can probably easily see where this is a great enhancement when it comes to applications such as databases that use multiple different storage volumes at different times, for example, data and logs.
If you create a snapshot of both volumes at different times, the application may not behave consistently when performing a disaster recovery. This is something to keep an eye on moving forward.
Prevent unauthorized volume mode conversion during volume restore
Another VolumeSnapshot-focused feature, this time graduating to beta and adds a layer of security to the VolumeSnapshot feature which has been around since 1.20. This feature will prevent the unauthorised conversion of the volume mode. This seems to be a pre-emptive feature around CVEs and other vulnerabilities that may arise in the future.
That is only a handful of the storage-focused and overall Kubernetes features in the 1.27 release. If this is useful let me know and we can keep on doing these when 1.28 and the next releases land. Also, let me know what your most interesting feature is from the list of features.
You can read in more detail the other enhancements for the release
Cloud Costs & Control
I want to jump out of the Kubernetes-focused content and step back into the wider Cloud Computing sphere, one of the common talk tracks that I am hearing daily is all about cloud costs and even sometimes this cloud cost burden forces people to bring back their workloads into a data centre… Which to me seems crazy!
Rehosting to Re-architecting
But when companies think they are in the cloud when they move up their VMs to AWS or one of the other hyper scalers, this is where the costs can easily spiral out of control. As technologists in this space, we have to encourage these companies and the industry in general that you cannot stop at the VM, you will always have some VMs either on-prem or in the cloud but you have to consider other cloud-based services as the next step. This is my first area of focus and when I get some free time, my summer project is going to be looking into these steps.
How can we help companies and teams to reimagine how their applications and services are architected and developed using Cloud-Native features?
Olden but Golden article from AWS
Tagging: The Foundation of Control in the Cloud
I am a big fan of tagging our workloads, data and environment.
When it comes to the Cloud you have the same tagging capability and my advice is that you should be using it to ensure the visibility of public cloud resources and their associated costs. It also allows for us to define our data protection policies based on these tags.
Most companies are going to need that help in implementing and enforcing tagging policies across most likely a hybrid environment, this is where I intend to spend some time in the summer running through this workflow.
Cloud cost optimisation
Cloud cost optimisation is a combination of strategies, techniques, best practices, and tools that not only help reduce cloud costs but also maximize the business value of using the cloud. We should be helping the industry via thought leadership and tooling that makes this process easier vs making the outcome of high bills a reversion of having to bring all that data out of the cloud and lose the benefits of the cloud in the long run.
As mentioned several times, I have a lot of thoughts on this area and it is important not only to Kubernetes administrators but also to Cloud Engineers, of which I strongly believe we are going to see the evolution continue of the systems administrator. I mentioned this a few weeks ago,
DeveloperWeek Europe
The final thing I want to add before signing off for a week in Amsterdam for KubeCon is I have been invited to speak at DeveloperWeek Europe in a couple of weeks’ time. The topic is around #90DaysOfDevOps and I have 25 minutes to cover all of those days…
DeveloperWeek Europe 2023 brings you cutting-edge developer learning from industry leaders. Discover the newest platforms, languages, technologies, and tools covering API & Microservices, AI & ML, Javascript, DevOps, Containers, Kubernetes, Cloud Computing, Dev Leadership, and more!
As a speaker, the organisers have been kind enough to provide me with a free speaker pass for the community and I would love for you all to come and join in and learn something from the two days. If you use this link below you will be able to register for FREE
There are lots of great sessions over the two-day event and that alone can seem pretty overwhelming, but my suggestion is to take a look at the “DevOps & Security” and “Containers & Kubernetes” tracks in the schedule to get started.
https://www.developerweek.com/europe/conference/
Also, I know it is called “Developer Week” but do not feel that you have to be a developer! One thing I have found over the last 3 years of being in the Cloud-Native space is that everyone is a developer!
Ok, that is probably enough for one post, over the next few weeks we will start looking at some upcoming events. KubeHuddle and VeeamON are also appearing over the mountain and we will get to them in a few weeks’ time.
