text stringlengths 0 1.8k |
|---|
**Gerhard Lazu:** Not yet. \[laughter\] |
**Jerod Santo:** Not yet. \[laughing out loud\] He just added that to his list of things to do. |
**Adam Stacoviak:** That's some serious dedication. |
**Jerod Santo:** Don't give him anything else to do, Adam. |
**Adam Stacoviak:** I'm just thinking like - the logistics of doing that. That's being on the phone to order it, that's deciding to pay for it. That's one more line item on the budget, so to speak. That's somebody coming in your house, new hole, new fiber, new equipment. At least you're getting to use that LAN fill ove... |
**Gerhard Lazu:** I do actually, yeah. I do. Not load balancing yet, but I'm working towards it. |
**Adam Stacoviak:** I’m sure you’ll be -- yeah. |
**Gerhard Lazu:** \[laughs\] |
**Jerod Santo:** Alright, we’ve got to reel this in. What's the summary here, Gerhard? What's the takeaway from this? |
**Gerhard Lazu:** The summary is that now I have two LAN connections |
**Jerod Santo:** \[laughing out loud\] You've already said that part. What's the takeaway? |
**Adam Stacoviak:** What's the takeaway here? |
**Gerhard Lazu:** You need two of each. \[laughing out loud\] Except your life. You only want one of those. |
**Jerod Santo:** There you go. So I think we should do GitPod and Codespaces. \[Laughter\] |
**Gerhard Lazu:** \[laughs\] Of course. |
**Adam Stacoviak:** Yes, because you never know. |
**Jerod Santo:** Kubernetes AND Fly.io, AND Render. That’s how we roll. |
**Adam Stacoviak:** Well, I can agree with the N-plus. I mean that is smart. I mean, you can never have enough. That was actually coined best in the movie Contact. Anybody remember that? Why build one when you can build two? |
**Gerhard Lazu:** I think I've had enough fun... \[laughter\] |
• Kaizen discussion and changelog compilation |
• GitOps implementation challenges |
• Using Honeycomb for question asking |
• Jared's work on moving static files to S3-like object store |
• Christmas gifts and upcoming episode themes |
• Importance of GitHub Codespaces integration |
• Backstage 20 release and Changelog 459 relation |
• Partnership announcements (Fastly, LaunchDarkly, Linode) |
• Fly platform promotion for global app deployment |
• The conversation revolves around an incident that occurred and how it has been addressed |
• The incident is related to a GitOps approach where unhealthy pods were put back into service using the "latest" tag |
• The speaker explains that the latest tag refers to the most recent version, which can be problematic if the latest version is broken or outdated |
• A tradeoff was made in the past by allowing corners to be cut for efficiency, but it failed once and now needs to be revisited |
• The conversation turns to a need to revisit the GitOps approach and consider using fixed versions instead of "latest" to avoid similar incidents |
• The speaker reflects on how they learned about the incident through investigation and analysis of the manifest |
• They discuss the importance of capturing specific version numbers for apps and containers to avoid infinite loops or failures. |
• Difficulty in handling unfinished TV shows and similar content |
• Discussion of GitOps and its implementation |
• Current lack of proper GitOps configuration in the system |
• Need to use GitOps properly with tools like Argo CD |
• Reviewing resources for learning about GitOps, including GitOps.org and WeWorks' website |
• Understanding of push-based and pull-based deployment models |
• Importance of following a standardized flow for deployments to prevent issues like using "latest" when it's broken |
• Capturing production for recovery from disaster |
• Importance of specific SHA (version) tagging in container images |
• Incident where a "disaster" occurred due to a broken "latest" tag and continuity not being in place |
• The role of caching and CDN in minimizing the impact of the incident |
• Integration with Honeycomb for enhanced visibility into app behavior and interactions |
• Resolving an issue related to slow MP3 requests using pull request/issue 383. |
• Integration with Honeycomb provides detailed insights into CDN performance |
• Observability features allow for slicing and dicing of data by various criteria, including top URLs, browser, user agent, country, and city |
• Cache status and hit/miss rates can be viewed in detail, as well as audio request breakdowns |
• Derived queries enable exploration of specific issues and root causes |
• Slow requests and their impact on website performance are identified, with an example of a GIF taking 1.4 minutes to load due to its large size and need for travel to the data center |
• Discussion of a large GIF file being served from Newark to Hong Kong and taking a long time to load |
• Use of lazy loading to improve user experience by allowing content to be accessed while the image is still loading |
• Issues with cache misses on Fastly CDN, including expired content not being cached indefinitely |
• Limitations of current caching setup, including headers asking for content to be kept in CDN for a few weeks but it may expire when requested again |
• Discussion of paying for caching service to keep all content cached globally |
• Assessment of total assets and weight of content to determine feasibility of caching everything |
• Discussion of latency and tail latency issues |
• Idea to serve static content from CDN instead of own infrastructure |
• Law of diminishing returns and slow clients consuming resources |
• Need for optimization in serving non-cached content from CDN |
• Requesting Fastly to cache content indefinitely |
• Imposter syndrome due to feeling like not using CDN correctly |
• The speaker is frustrated with the performance of their CDN, specifically Fastly |
• They discuss how long-term content is not being cached as expected |
• They mention that they are a media company with static content that doesn't change often |
• They express confusion about why the CDN is not working as intended despite following best practices |
• They suggest that there may be issues within the CDN itself or its configuration that need to be addressed |
• They propose reaching out to Fastly for further assistance and collaboration. |
• Discussion of a high number of misses in MP3 file delivery |
• Importance of data visibility and the impact of having "hard facts" with Honeycomb integration |
• Conversation about Fastly's functionality and how it handles misses |
• Reference to LaGuardia and Hong Kong locations, possibly related to edge caching or CDN configuration |
• Explanation of Incident.io and its role in incident management within Slack |
• Promotion of Raygun for performance monitoring and user insights |
• Personal anecdote about setting up new M1 Macs |
• Discussing challenges with setting up M1 Mac for changel.com development environment |
• Considering alternatives to Docker and GitHub Code Spaces |
• Wanting to run own infrastructure on Equinix Metal or Linode |
• Concerns about cost and simplicity of GitHub Code Spaces |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.