Hi! I hope you can give me some advice. I have all that infrastructure, including a jenkins server, where we run our acceptance tests pipelines. The thing is that we want to have a status page, because when a set of tests fail, we want to know if there is an issue with the infrastructure, or even if this is a tests that failed before. We have grafana dashboards, with all those nice graphs, but I need something more simple and direct, a list of "things" and then a 👍 👎 next to it. Any advice on how to do this? Should I just do it with grafana itself? Thanks in advance

Hello @Shai Almog. You wrote quite a bit about production problems. What are your top tips for dealing with production issues?

Hi. Your career seems pretty varied in the experiences. What did you take from the different stages that's still applicable today? Are 90's problems still relevant today?

Anyone here missed @Chris Tomkins’ webinar on how to secure your Kubernetes Container Network Interface (CNI) with Calico? Good news, Chris is also giving a talk at PlatformCon about the usage of open-source software to secure cloud-native applications! Go check it out! ☝️

It can be hard to build front ends for internal apps for non devs wonder if something like https://www.appsmith.com/blog/the-modern-stack-to-build-internal-tools-supabase-appsmith-n8n can be used ( for those of us who dont have dedicated front end teams 😂 ) , specially now that supabase has functions powered by deno https://supabase.com/edge-functions

Good news! We’re back with AMA sessions 🚀 Ian Eyberg, CEO at NanoVMs, promised to answer anything that you are interested about cloud complexity and Unikernels before his webinar on 24th of May. Date: May 17 Time: 5pm (GMT+2) / 8am (GMT-7) Channel: #platform-tech See you there? 😜

Looks like weaveworks have opened up their liquid metal repo https://github.com/weaveworks-liquidmetal / https://www.weave.works/blog/multi-cluster-kubernetes-on-microvms-for-bare-metal since were talking about unikernels 😄

At Kubecon 2021, Razorpay presented its open-source solution for improving developer productivity. The session at PlatformCon by @Srinidhi S and @Venkat V will build on that and shed 💡 on extending a pure Kubernetes native development environment with remote debugging capabilities, data capsules and smart scaling.

How’s Viktor Farcic keeping up with everything he does?! He is a developer advocate at Upbound, very active onTwitter, runs his own Youtube channel, creates online courses, writes books, etc. But more importantly, he’s a PlatformCon speaker 😛latformcon:. Viktor will walk us through the steps of how to build an IDP with Backstage, Crossplane and Argo CD.

Keeping my introduction as background, could any of you help me with any metrics that would be useful to track team performance? For instance - https://thenewstack.io/googles-formula-for-elite-devops-performance/ I came across the DORA metrics. Any that your teams track and how they help? I’d love if we could chat more on this thread, feel free to DM too 🙂

Hey all! @Ian Eyberg is ready to answer anything you’re interested about cloud complexity and Unikernels. You now have 30min, so use this time wisely 😉

Hi @Ian Eyberg, how does Nanos compare to other unikernels?

Does Nanos run under kubernetes?

sorry running late

hi @Damien Ricci - good question - there's a nice little comparison box here https://nanos.org/faq but some of the larger differences is that we keep the kernel<>user switch for security reasons, we also can load raw elf binaries (native linux format) which means we can basically run any linux app without modifications if they fit into the unikernel architecture

@Naya Kumar yes, you definitely can but it isn't really advised - i wrote a short tutorial on this a while back ago https://hackernoon.com/running-linux-applications-as-unikernels-with-k8s-gl2p3yfr

the reason is that kubernetes was built to run containers not virtual machines and since most people are running on the public cloud (which is inherently built on virtual machines) it presents multiple layers degrading security and performance

when we deploy to the cloud we create amis out of your application - no k8s, no linux at all - when it boots up it's just your app

@Ian Eyberg How can I share unikernels with other folks?

@Dominic Walt best way is through https://repo.ops.city - we have the concept of 'packages' for applications that require it; some apps you might need to create a package around as they can be fairly simple - for instance both https://nanos.org and https://ops.city are go unikernels running on GCP but for something like say a python ML application - you'd probably want a package

to elaborate - we don't (today) do the layered filesystem approach, however, you can take an existing package and add to it as necessary - so you might take an elixir package and add your code to it or a jvm package and add code to it to create your own unikernel

@Ian Eyberg thank you for you time! See you at the webinar on May 24 🙂

Have you ever heard of the EKS Creation Engine (ECE) tool? It started as an internal project that rapidly matured based on feedback and was released as an open-source tool gaining some cult following. Jonathan Rau, CISO at Lightspin will give a walk-through of ECE and discuss: • The rationale of bucking trends • Iterating quickly on developer feedback • Why multi-optioned CLIs can be better than declarative language for IaC use cases All at PlatformCon 😛latformcon:.

Hi Everyone, 👋 What are messaging queues everyone is using and what are your experiences with those. I'm looking at exploring Kafka, Apache Airflow, RabbitMQ and anything that you all suggest. Our requirement is an MQ that can: 1) Scale 2) Is Pub-Sub model based with Topic (like) support 3) Good Java library support 4) Works well for multi-region and multi-cloud setups. 5) Works well in failovers and has HA built-in. 6) Provides admission control for apps. For e.g., App A can read + write, App B can only Read, App C can only write etc.

Hi all - If anyone is interested, we open sourced our SQL-on-JSON engine https://github.com/SnellerInc/sneller. You can use it to run standard SQL on TBs of semi-structured and deeply nested JSON data that you may be gathering from your Ops pipelines. Hope it's interesting and we look forward to any/all feedback!

Familiar with minimum viable security (MVS)? Aviram Shmueli, CPO & Co-Founder at Jit, will walk us through the MVS mindset and how it's applied to cloud-native apps. He will cover the 6 critical security controls that must be integrated into your workflow as part of secure cloud-native operations and CI/CD.

Hi, is anyone using backstage.io? If so, what are your thoughts on the tool?

At Pangea, the platform team tries to optimize the speed at which incremental changes land into production. @Akshay Dongaonkar will give a talk on how a subset of the infrastructure is provisioned every time a service author opens up a merge request using Gitlab and Terraform. Interested? 🤔 Prepare some questions for Akshay for the live Q&A session. He will be there on the 10th of June! 👈

Building software DAGs with the power of BuildKit and Cue. @Erik Sipsma, Senior Software Engineer at Dagger will give a talk about ways of taming tangled webs through a few tools centered around the concept of directed acyclic graphs (DAGs). He will cover: • BuildKit is a low-level DAG execution engine that takes care of hard problems like scheduling, caching and exporting data. • Cue is a configuration language that allows independent systems making up software delivery DAGs to be glued together in a robust, type-checked manner. • These tools can be used independently, but through Dagger they can also be combined into a powerful framework for orchestrating complex software DAGs.