text stringlengths 0 1.8k |
|---|
**Gerhard Lazu:** Right. |
**Muvaffak Onuş:** So you would have the cluster that is managed in one namespace, maybe like Changelog system, with its own claim, claim is what we call similar to PVC. So you would have that production cluster, but different teams or developers in their own namespace - they would refer to that central production clus... |
**Gerhard Lazu:** Yeah. |
**Muvaffak Onuş:** So it's about like in publishing a new API - instead of going through all the fields of the specific clouds, you would publish API, with the only difference that you want it to be configured. |
**Gerhard Lazu:** Okay. That is really cool. I can hardly wait to do that. That is seriously cool. Having all this stuff abstracted in a composition, to just capture what it means for the entire Changelog setup to come online, would be so amazing. |
The other thing which would be also amazing is to move Crossplane from being hosted on our cluster, to be hosted on Upbound cloud. Because the dream is there is a seed cluster somewhere, which is managed by someone else, in this case Upbound cloud. The Crossplane is there, we can define all the important stuff, and tha... |
Again, I don't wanna go too far with this idea, like blow your minds completely, but why doesn't it manage some Fly.io apps? Or why doesn't it manage maybe some DNS? Or why doesn't it manage other things from the seed cluster? Because right now, the external DNS is what we use in every cluster to manage its own DNS. An... |
Well, I'm really looking forward to what we'll do together next year, Muvaffak, with all this stuff. There's so many improvements which we can drive... I'm really keen on that. It's the first step. But you as a listener, what I would say is have a look at the provider Jet Linode in the Crossplane Contrib org, see if it... |
**Muvaffak Onuş:** Yeah, it was great working with you for the last couple of days to get all these things done. Yeah, I'm honored to be here. Happy Christmas. |
**Gerhard Lazu:** Thank you, Muvaffak. It's been my pleasure, thank you very much. See you next year! |
• CI/CD Lego set (PR395) for changelog.com |
• Continuous CPU profiling (PR396) |
• Auto-restoring Kubernetes clusters (PR399) |
• Dagger, a universal deployment engine introduced in episode 23 |
• Implementation and challenges of integrating Dagger with changelog's infrastructure |
• Benefits of using Dagger for CI/CD pipelines |
• Interactions between Q and BuildKit and how they're applied to BuildKit states |
• Improvements in the new pipeline for compiling dependencies and running tests in parallel |
• Use of caching mechanisms and ephemeral databases |
• Changes to the API to make it more explicit and intuitive |
• Integration with GitHub Actions and its potential use in the pipeline |
• Overview of Dagger's current state and improvements being made |
• Pipeline execution in parallel |
• Caching of steps to speed up pipeline execution |
• Dagger's ability to use caching for faster builds |
• Comparison of Dagger and Docker build times |
• OpenTracing integration for visualization of pipeline execution |
• Flexibility to run pipelines on any CI setup, including GitHub Actions and CircleCI |
• Demonstration of Dagger's cached run and its performance |
• Discussion on using BuildKit to enable parallel execution of fine-grained steps |
• Explanation of how GitHub Actions integration works with Dagger |
• Breakdown of a specific GitHub Actions config file, including environment variables, jobs, and actions used |
• Description of using Tailscale Tunnel for remote Docker access |
• Discussion on committing secrets and using SOPS for encryption |
• Transitioning from "environment" to "plan" or "DAG" |
• Using CI package for tasks and automation |
• Docker-based workflow with encrypted credentials |
• Providing inputs for environment/plan/DAG, including source code and dependencies |
• Explanation of queue configuration language and its features (schema definition, data validation) |
• DAG or pipeline stages (app image, test container, etc.) |
• Depths compile as a way to describe and apply structures in other places |
• Discussion of the advantages of Queue over other configuration management tools like YAML and Helm |
• Explanation of how Queue's schema definition feature can define the shape of a particular configuration, including constraints on different fields |
• Example use case for Kubernetes deployments with CPU field constraints |
• Description of how Queue's compiler errors steer users in the right direction |
• Discussion of potential future developments in the Queue community, including a language server |
• Mention of upcoming changes with Europa that will simplify and improve the user experience |
• Profiling has been a key tool for developers since software engineering began |
• Traditional profiling was expensive and only done when necessary |
• Sampling profiling introduced a more efficient method, recording stack traces at statistically significant intervals (e.g., 100 times per second) |
• This allows for detailed analysis of CPU usage and optimization opportunities |
• Parca is an implementation that provides near real-time flame graphs and comparative CPU profiles for Kubernetes workloads |
• It uses EVPF technology to minimize overhead and record stack traces directly in the kernel |
• Benefits include saving money, improving performance, and understanding system bottlenecks. |
• The Parca agent sends CPU profiles to the Parca server for analysis and visualization. |
• The PPF (Profiling Format) standard is used for profiling data, allowing integration with other tooling and workflows. |
• The server ingests CPU profiles from the agent and allows downloading in PPF format for external use. |
• The Parca UI displays flame graphs, showing cumulative values of CPU usage for each span. |
• Garbage collection can be a significant contributor to CPU spikes due to memory allocation and deallocation. |
• Continuous profiling and symbolization can contribute to garbage collection spikes. |
• Optimizations such as buffer reuse can help reduce memory allocations. |
• The discussion begins with examining a Postgres database and its compiled binary, observing that it only shows memory addresses due to debug information being intentionally removed for size optimization. |
• A workaround is mentioned using Debug Info D, a project hosting servers for on-demand debug symbol retrieval. |
• Parker doesn't currently support this feature, but the developers are working on implementing it. |
• The discussion moves on to examining an Erlang VM and its interpreted code, noting that it has a just-in-time compiler. |
• A comparison is made with other dynamic languages and virtual machines, highlighting the need for runtime-specific implementation in Parker. |
• The use of perf maps in Erlang is mentioned as a potential solution for symbolizing memory addresses on the fly. |
• There are issues with implementing this feature for Erlang, but support is expected to be added in the future. |
• The discussion concludes with an overview of the compare view in Parker, which allows users to compare two profiles side by side. |
• Observations of CPU cycles and stack traces |
• Analysis of performance metrics using a diff function |
• Discussion of memory address symbolization |
• Investigation into a memory address with unknown origin |
• Use of Linux procfs to inspect binary code memory mapping |
• Attempt to identify the source of an unknown memory address |
• Discussion of a table that shows executable code coming from the binary |
• Explanation of how the stack can sometimes be too tall when retrieving stack trace snapshots from eBPF |
• Theoretical possibility of executing code on the stack in Erlang VM |
• Proposal to ask Lucas Larson for expertise on the matter |
• Mention of pull request 396 and a plan to follow up on it |
• Review of Parca discussion and agreement to fix issue R |
• Plan to use cross-plane to provision Linode Kubernetes cluster using TerraJet tool |
• Discussion of testing generated provider with cross-plane control plane or local kind cluster |
• Proposal to start with production setup, including installing cross-plane in the production environment |
• discussion about providing a kubeconfig to access a cluster |
• mention of TerraJet and its use in generating providers for crossplane |
• confirmation that version 1.3 of crossplane is sufficient for using TerraJet |
• demonstration of connecting to a system using TerraJet |
• discussion of the TerraJet generation process and collaboration with other team members |
• introduction of ProviderJet, a template for creating new repositories |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.