Advertise on podcast: Ship It Weekly - DevOps, SRE, and Platform Engineering News

This week on Ship It Weekly, Brian hits four “control plane + trust boundary” stories where the glue layer becomes the incident. Azure had a platform incident that impacted VM management operations across multiple regions. Your app can be up, but ops is degraded. GitHub is pushing Agent HQ (Claude + Codex in the repo/CI flow), and Actions added a case() function so workflow logic is less brittle. MCP is becoming platform plumbing: Miro launched an MCP server and Kong launched an MCP Registry. Links Azure status incident (VM service management issues) https://azure.status.microsoft/en-us/status/history/?trackingId=FNJ8-VQZ GitHub Agent HQ: Claude + Codex https://github.blog/news-insights/company-news/pick-your-agent-use-claude-and-codex-on-agent-hq/ GitHub Actions update (case() function) https://github.blog/changelog/2026-01-29-github-actions-smarter-editing-clearer-debugging-and-a-new-case-function/ Claude Opus 4.6 https://www.anthropic.com/news/claude-opus-4-6 How Google SREs use Gemini CLI https://cloud.google.com/blog/topics/developers-practitioners/how-google-sres-use-gemini-cli-to-solve-real-world-outages Miro MCP server announcement https://www.businesswire.com/news/home/20260202411670/en/Miro-Launches-MCP-Server-to-Connect-Visual-Collaboration-With-AI-Coding-Tools Kong MCP Registry announcement https://konghq.com/company/press-room/press-release/kong-introduces-mcp-registry GitHub Actions hosted runners incident thread https://github.com/orgs/community/discussions/186184 DockerDash / Ask Gordon research https://noma.security/blog/dockerdash-two-attack-paths-one-ai-supply-chain-crisis/ Terraform 1.15 alpha https://github.com/hashicorp/terraform/releases/tag/v1.15.0-alpha20260204 Wiz Moltbook write-up https://www.wiz.io/blog/exposed-moltbook-database-reveals-millions-of-api-keys Chainguard “EmeritOSS” https://www.chainguard.dev/unchained/introducing-chainguard-emeritoss More episodes + details: https://shipitweekly.fm

This week on Ship It Weekly, Brian looks at four “glue failures” that can turn into real outages and real security risk. We start with CodeBreach: AWS disclosed a CodeBuild webhook filter misconfig in a small set of AWS-managed repos. The takeaway is simple: CI trigger logic is part of your security boundary now. Next is the Bazel TLS cert expiry incident. Cert failures are a binary cliff, and “auto renew” is only one link in the chain. Third is Helm chart reliability. Prequel reviewed 105 charts and found a lot of demo-friendly defaults that don’t hold up under real load, rollouts, or node drains. Fourth is n8n. Two new high-severity flaws disclosed by JFrog. “Authenticated” still matters because workflow authoring is basically code execution, and these tools sit next to your secrets. Lightning round: Fence, HashiCorp agent-skills, marimo, and a cautionary agent-loop story. Links AWS CodeBreach bulletin https://aws.amazon.com/security/security-bulletins/2026-002-AWS/ Wiz research https://www.wiz.io/blog/wiz-research-codebreach-vulnerability-aws-codebuild Bazel postmortem https://blog.bazel.build/2026/01/16/ssl-cert-expiry.html Helm report https://www.prequel.dev/blog-post/the-real-state-of-helm-chart-reliability-2025-hidden-risks-in-100-open-source-charts n8n coverage https://thehackernews.com/2026/01/two-high-severity-n8n-flaws-allow.html Fence https://github.com/Use-Tusk/fence agent-skills https://github.com/hashicorp/agent-skills marimo https://marimo.io/ Agent loop story https://www.theregister.com/2026/01/27/ralph_wiggum_claude_loops/ Related n8n episodes: https://www.tellerstech.com/ship-it-weekly/n8n-critical-cve-cve-2026-21858-aws-gpu-capacity-blocks-price-hike-netflix-temporal/ https://www.tellerstech.com/ship-it-weekly/n8n-auth-rce-cve-2026-21877-github-artifact-permissions-and-aws-devops-agent-lessons/ More episodes + details: https://shipitweekly.fm

This is a guest conversation episode of Ship It Weekly (separate from the weekly news recaps). In this Ship It: Conversations episode I talk with Austin Reed from horizon.dev about AI and automation for small and mid-sized businesses, and what actually works once you leave the demo world. We get into the most common automation wins he sees (sales and customer service), why a lot of projects fail due to communication and unclear specs more than the tech, and the trap of thinking “AI makes it cheap.” Austin shares how they push teams toward quick wins first, then iterate with prototypes so you don’t spend $10k automating a thing that never even happens. We also talk guardrails: when “human-in-the-loop” makes sense, what he avoids automating (finance-heavy logic, HIPAA/medical, government), and why the goal is usually leverage, not replacing people. On the dev side, we nerd out a bit on the tooling they’re using day to day: GPT and Claude, Cursor, PR review help, CI/CD workflows, and why knowing how to architect and validate output matters way more than people think. If you’re a DevOps/SRE type helping the business “do AI,” or you’re just tired of automation hype that ignores real constraints like credentials, scope creep, and operational risk, this one is very much about the practical middle ground. Links from the episode: Austin on LinkedIn: https://www.linkedin.com/in/automationsexpert/ horizon.dev: horizon.dev YouTube: https://www.youtube.com/@horizonsoftwaredev Skool: https://www.skool.com/automation-masters If you found this useful, share it with the person on your team who keeps saying “we should automate that” but hasn’t dealt with the messy parts yet. More information on our website: https://shipitweekly.fm

This week on Ship It Weekly, Brian looks at three different versions of the same problem: systems are getting faster, but human attention is still the bottleneck. We start with curl shutting down their bug bounty program after getting flooded with low-quality “AI slop” reports. It’s not a “security vs maintainers” story, it’s an incentives and signal-to-noise story. When the cost to generate reports goes to zero, you basically DoS the people doing triage. Next, AWS improved RDS Blue/Green Deployments to cut writer switchover downtime to typically ~5 seconds or less (single-region). That’s a big deal, but “fast switchover” doesn’t automatically mean “safe upgrade.” Your connection pooling, retries, and app behavior still decide whether it’s a blip or a cascade. Third, Amazon ECR added cross-repository layer sharing. Sounds small, but if you’ve got a lot of repos and you’re constantly rebuilding/pushing the same base layers, this can reduce storage duplication and speed up pushes in real fleets. Lightning round covers a practical Kubernetes clientcmd write-up, a solid “robust Helm charts” post, a traceroute-on-steroids style tool, and Docker Kanvas as another signal that vendors are trying to make “local-to-cloud” workflows feel less painful. We wrap with Honeycomb’s interim report on their extended EU outage, and the part that always hits hardest in long incidents: managing engineer energy and coordination over multiple days is a first-class reliability concern. Links from this episode curl bug bounties shutdown https://github.com/curl/curl/pull/20312 RDS Blue/Green faster switchover https://aws.amazon.com/about-aws/whats-new/2026/01/amazon-rds-blue-green-deployments-reduces-downtime/ ECR cross-repo layer sharing https://aws.amazon.com/about-aws/whats-new/2026/01/amazon-ecr-cross-repository-layer-sharing/ Kubernetes clientcmd apiserver access https://kubernetes.io/blog/2026/01/19/clientcmd-apiserver-access/ Building robust Helm charts https://www.willmunn.xyz/devops/helm/kubernetes/2026/01/17/building-robust-helm-charts.html ttl tool https://github.com/lance0/ttl Docker Kanvas (InfoQ) https://www.infoq.com/news/2026/01/docker-kanvas-cloud-deployment/ Honeycomb EU interim report https://status.honeycomb.io/incidents/pjzh0mtqw3vt SRE Weekly issue #504 https://sreweekly.com/sre-weekly-issue-504/ More episodes + details: https://shipitweekly.fm

This week on Ship It Weekly, the theme is simple: the automation layer has become a control plane, and that changes how you should think about risk. We start with n8n’s latest critical vulnerability, CVE-2026-21877. This one is different from the unauth “Ni8mare” issue we covered in Episode 12. It’s authenticated RCE, which means the real question isn’t only “is it internet exposed,” it’s who can log in, who can create or modify workflows, and what those workflows can reach. Takeaway: treat workflow automation tools like CI systems. They run code, they hold credentials, and they can pivot into real infrastructure. Next is GitHub’s new fine-grained permission for artifact metadata. Small change, big least-privilege implications for Actions workflows. It’s also a good forcing function to clean up permission sprawl across repos. Third is AWS’s DevOps Agent story, and the best part is that it’s not hype. It’s a real look at what it takes to operationalize agents: evaluation, observability into tool calls/decisions, and control loops with brakes and approvals. Prototype is cheap. Reliability is the work. Lightning round: GitHub secret scanning changes that can quietly impact governance, a punchy Claude Code “guardrails aren’t guaranteed” reminder, Block’s Goose as another example of agent workflows getting productized, and OpenCode as an “agent runner” pattern worth watching if you’re experimenting locally. Links n8n CVE-2026-21877 (authenticated RCE) https://thehackernews.com/2026/01/n8n-warns-of-cvss-100-rce-vulnerability.html?m=1 Episode 12 (n8n “Ni8mare” / CVE-2026-21858) https://www.tellerstech.com/ship-it-weekly/n8n-critical-cve-cve-2026-21858-aws-gpu-capacity-blocks-price-hike-netflix-temporal/ GitHub: fine-grained permission for artifact metadata (GA) https://github.blog/changelog/2026-01-13-new-fine-grained-permission-for-artifact-metadata-is-now-generally-available/ GitHub secret scanning: extended metadata auto-enabled (Feb 18) https://github.blog/changelog/2026-01-15-secret-scanning-extended-metadata-to-be-automatically-enabled-for-certain-repositories/ Claude Code issue thread (Bedrock guardrails gap) https://github.com/anthropics/claude-code/issues/17118 Block Goose (tutorial + sessions/context) https://block.github.io/goose/docs/tutorials/rpi https://block.github.io/goose/docs/guides/sessions/smart-context-management OpenCode https://opencode.ai More episodes + details: https://shipitweekly.fm

This is a guest conversation episode of Ship It Weekly (separate from the weekly news recaps). In this Ship It: Conversations episode I talk with Gracious James Eluvathingal about TARS, his “human-in-the-loop” fixer bot wired into CI/CD. We get into why he built it in the first place, how he stitches together n8n, GitHub, SSH, and guardrailed commands, and what it actually looks like when an AI agent helps with incident response without being allowed to nuke prod. We also dig into rollback phases, where humans stay in the loop, and why validating every LLM output before acting on it is the single most important guardrail. If you’re curious about AI agents in pipelines but hate the idea of a fully autonomous “ops bot,” this one is very much about the middle ground: segmenting workflows, limiting blast radius, and using agents to reduce toil instead of replace engineers. Gracious also walks through where he’d like to take TARS next (Terraform, infra-level decisions, more tools) and gives some solid advice for teams who want to experiment with agents in CI/CD without starting with “let’s give it root and see what happens.” Links from the episode: Gracious on LinkedIn: https://www.linkedin.com/in/gracious-james-eluvathingal TARS overview post: https://www.linkedin.com/posts/gracious-james-eluvathingal_aiagents-devops-automation-activity-7391064503892987904-psQ4 If you found this useful, share it with the person on your team who’s poking at AI automation and worrying about guardrails. More information on our website: https://shipitweekly.fm

This week on Ship It Weekly, Brian’s theme is basically: the “automation layer” is not a side tool anymore. It’s part of your perimeter, part of your reliability story, and sometimes part of your budget problem too. We start with the n8n security issue. A lot of teams use n8n as glue for ops workflows, which means it tends to collect credentials and touch real systems. When something like this drops, the right move is to treat it like production-adjacent infra: patch fast, restrict exposure, and assume anything stored in the tool is high value. Next is AWS quietly raising prices on EC2 Capacity Blocks for ML. Even if you’re not a GPU-heavy shop, it’s a useful signal: scarce compute behaves like a market. If you do rely on scheduled GPU capacity, it’s time to revisit forecasts and make sure your FinOps tripwires catch rate changes before the end-of-month surprise. Third is Netflix’s write-up on using Temporal for reliable cloud operations. The best takeaway is not “go adopt Temporal tomorrow.” It’s the pattern: long-running operational workflows should be resumable, observable, and safe to retry. If your critical ops are still bash scripts and brittle pipelines, you’re one transient failure away from a very dumb day. In the lightning round: Kubernetes Dashboard getting archived and the “ops dependencies die” reality check, Docker pushing hardened images as a safer baseline and Pipedash. Links SRE Weekly issue 504 (source roundup) https://sreweekly.com/sre-weekly-issue-504/ n8n CVE (NVD) https://nvd.nist.gov/vuln/detail/CVE-2026-21858 n8n community advisory https://community.n8n.io/t/security-advisory-security-vulnerability-in-n8n-versions-1-65-1-120-4/247305 AWS price increase coverage (The Register) https://www.theregister.com/2026/01/05/aws_price_increase/ Netflix: Temporal powering reliable cloud operations https://netflixtechblog.com/how-temporal-powers-reliable-cloud-operations-at-netflix-73c69ccb5953 Kubernetes SIG-UI thread (Dashboard archiving) https://groups.google.com/g/kubernetes-sig-ui/c/vpYIRDMysek/m/wd2iedUKDwAJ Kubernetes Dashboard repo (archived) https://github.com/kubernetes/dashboard Pipedash https://github.com/hcavarsan/pipedash Docker Hardened Images https://www.docker.com/blog/docker-hardened-images-for-every-developer/ More episodes and more details on this episode can be found on our website: https://shipitweekly.fm

This is a guest conversation episode of Ship It Weekly (separate from the weekly news recaps). I sat down with Danny Teller, a DevOps Architect and Tech Lead Manager at Tipalti, to talk about internal developer platforms and the reality behind “just set up a developer portal.” We get into Backstage versus internal IDPs, why adoption is the real battle, and why platform/DevEx maturity matters more than whatever tool you pick. What we covered Backstage vs internal IDPs Backstage is a solid starting point for a developer portal, but it doesn’t magically create standards, ownership, or platform maturity. We talk about when Backstage fits, and when teams end up building internal tooling anyway. DevEx muscle (the make-or-break) Danny’s take: the portal UI is the easy part. The hard part is the ongoing work that makes it useful: paved roads, sane defaults, support, and keeping the catalog/data accurate so engineers trust it. Where teams get burned Common failure mode: teams ship a portal first, then realize they don’t have the resourcing, ownership, or workflows behind it. Adoption fades fast if the portal doesn’t remove real friction. A build vs buy gut check We walk through practical signals that push you toward open source Backstage, a managed Backstage offering, or a commercial portal. We also hit the maintenance trap: if you build too much, you’ve created a second product. Links and resources Danny Teller's Linkedin: https://www.linkedin.com/in/danny-teller/ matlas — one CLI for Atlas and MongoDB: https://github.com/teabranch/matlas-cli Backstage: https://backstage.io/ Roadie (managed Backstage): https://roadie.io/ Port: https://www.port.io/ Cortex: https://www.cortex.io/ OpsLevel: https://www.opslevel.com/ Atlassian Compass: https://www.atlassian.com/software/compass Humanitec Platform Orchestrator: https://humanitec.com/products/platform-orchestrator Northflank: https://northflank.com/ If you enjoyed this episode Ship It Weekly is still the weekly news recap, and I’m dropping these guest convos in between. Follow/subscribe so you catch both, and if this was useful, share it with a platform/devex friend and leave a quick rating or review. It helps more than it should. Visit our website at https://www.shipitweekly.fm

This week on Ship It Weekly, Brian kicks off the new year with one theme: automation is getting faster, and that makes blast radius and oversight matter more than ever. We start with Cloudflare’s “fail small” mindset. The core idea is simple: big outages usually come from correlated failure, not one box dying. If a bad change lands everywhere at once, you’re toast. “Fail small” is about forcing problems to stay local so you can stop the bleeding before it becomes global. Next is Pulumi’s push to be the control plane for all your IaC, including Terraform and HCL. The interesting part isn’t syntax wars. It’s the workflow layer: approvals, policy enforcement, audit trails, drift, and how teams standardize without signing up for a multi-year rewrite. Third is Meta’s DrP, a root cause analysis platform that turns repeated incident investigation steps into software. Even if you’re not Meta, the pattern is worth stealing: automate the first 10–15 minutes of your most common incident types so on-call is consistent no matter who’s holding the pager. In the lightning round: a follow-up on GitHub Actions direction (and a quick callback to Episode 6’s runner pricing pause), AWS ECR creating repos on push, a smarter take on incident metrics, Terraform drift visibility, and parallel “coding agent” workflows. We wrap with a human reminder about the ironies of automation: automation doesn’t remove responsibility, it moves it. Faster systems require better brakes, better observability, and easier rollback. Links from this episode SRE Weekly issue 503 (source roundup - CloudFlare) https://sreweekly.com/sre-weekly-issue-503/ Pulumi: all IaC, including Terraform and HCL https://www.pulumi.com/blog/all-iac-including-terraform-and-hcl/ Meta DrP: https://engineering.fb.com/2025/12/19/data-infrastructure/drp-metas-root-cause-analysis-platform-at-scale/ GitHub Actions: “Let’s talk about GitHub Actions” https://github.blog/news-insights/product-news/lets-talk-about-github-actions/ Episode 6 (GitHub runner pricing pause, Terraform Cloud limits, AI in CI) https://www.tellerstech.com/ship-it-weekly/github-runner-pricing-pause-terraform-cloud-limits-and-ai-in-ci/ AWS ECR: create repositories on push https://aws.amazon.com/about-aws/whats-new/2025/12/amazon-ecr-creating-repositories-on-push/ DriftHound https://drifthound.io/ Superset https://superset.sh/ More episodes + contact info, and more details on this episode can be found on our website: https://shipitweekly.fm

This is a guest conversation episode of Ship It Weekly (separate from the weekly news recaps). I sat down with Eric Paatey, a Cloud & DevOps Engineer who’s been transitioning from full-stack web development into cloud/devops, and building real skills through hands-on projects instead of just collecting tools and buzzwords. We talk about what that transition actually feels like, what’s helped most, and why you don’t need a rack of servers to learn DevOps. What we covered Eric’s path into DevOps How he moved from building web apps to caring about pipelines, infra, scalability, reliability, and automation. The “oh… code is only part of the job” moment that pushes a lot of people toward DevOps. The WHY behind DevOps Eric’s take: DevOps is mainly about breaking down silos and improving communication between dev, ops, security, and the business. We also hit the idea from The DevOps Handbook: small batches win. The bigger the release, the harder it is to recover when something breaks. Leveling up without drowning in tools DevOps has an endless tool list, so we talked about how to stay current without burning out. Eric’s recommendation: stay connected to the industry. Meet people, join user groups, go to events, and don’t silo yourself. The homelab mindset (and why simple is fine) Eric shared his “homelab on the go” setup and why the hardware isn’t the point. It’s about using a safe environment to build habits: automation, debugging, systems thinking, monitoring, breaking things, recovering, and improving the design. A practical first project for aspiring DevOps engineers We talked through a starter project you can actually show in interviews: Dockerize a simple app, deploy it behind an ALB, and learn basic networking/security along the way. You don’t need to understand everything on day one, but you do need to build things and learn what breaks. Agentic AI and guardrails We also touched on AI agents and MCPs, what they could mean for ops teams, and why you should not give agents full access to anything. Least privilege and policy guardrails matter, because “non-deterministic” and “prod permissions” is a scary combo. Links and resources Eric Paatey on LinkedIn: https://www.linkedin.com/in/eric-paatey-72a87799/ Eric’s website/portfolio: https://ericpaatey.com/ If you enjoyed this episode Ship It Weekly is still the weekly news recap, and I’m dropping these guest convos in between. Follow/subscribe so you catch both, and if this was useful, share it with a coworker or your on-call buddy and leave a quick rating or review. It helps more than it should. Visit our website at https://www.shipitweekly.fm

This week on Ship It Weekly, Brian looks at real platform engineering in the wild. We start with Cloudflare’s write-up on building an internal maintenance scheduler on Workers. It’s not marketing fluff. It’s “we hit memory limits, changed the model, and stopped pulling giant datasets into the runtime.” Next up: AWS databases are now available inside the Vercel Marketplace. This is a quiet shift with loud consequences. Devs can click-button real AWS databases from the same place they deploy apps, and platform teams still own the guardrails: account sprawl, billing/tagging, audit trails, region choices, and networking posture. Third story: TEAM (Temporary Elevated Access Management) for IAM Identity Center. Time-bound elevation with approvals, automatic expiry, and auditing. We cover how this fits alongside break-glass and why auto-expiry is the difference between least-privilege and privilege creep. Lightning round: GitHub Actions workflow page performance improvements, Lambda Managed Instances (slightly cursed but interesting), a quick atmos tooling blip, and k8sdiagram.fun for explaining k8s to humans. We close with Marc Brooker’s “What Now? Handling Errors in Large Systems” and the takeaway: error handling isn’t a local code decision, it’s architecture. Crashing vs retrying vs continuing only makes sense when you understand correlation and blast radius. shipitweekly.fm has links + the contact email. Want to be a guest? Reach out. And if you’re enjoying the show, follow/subscribe and leave a quick rating or review. It helps a ton. Links from this episode Cloudflare https://blog.cloudflare.com/building-our-maintenance-scheduler-on-workers/ AWS on Vercel https://aws.amazon.com/about-aws/whats-new/2025/12/aws-databases-are-available-on-the-vercel/ https://vercel.com/changelog/aws-databases-now-available-on-the-vercel-marketplace TEAM https://aws-samples.github.io/iam-identity-center-team/ https://github.com/aws-samples/iam-identity-center-team GitHub Actions https://github.blog/changelog/2025-12-22-improved-performance-for-github-actions-workflows-page/ Lambda Managed Instances https://docs.aws.amazon.com/lambda/latest/dg/lambda-managed-instances.html Atmos https://github.com/cloudposse/atmos/issues k8sdiagram.fun https://k8sdiagram.fun/ Marc Brooker https://brooker.co.za/blog/2025/11/20/what-now.html

This is a Ship It Weekly conversation episode. The weekly news recaps are still weekly. These interviews drop in between when I find someone worth talking to and the convo feels useful. In this episode I’m joined by Mazharul “Maz” Islam (DevOps with Maz). Maz is a UK-based DevOps Engineer who shares practical, real-world DevOps content on YouTube and LinkedIn. We talk about the stuff that actually matters when you’re building systems, running infrastructure, owning reliability, and living in on-call. We hit three big things: the importance of understanding the WHY behind DevOps (not just the tools), how to upskill and keep up with the industry without burning out, and what the agentic AI era might look like for DevOps, SRE, and platform engineering teams. We also touch on MCPs and AI agents, and what “leveling up” looks like for companies that want to move faster without breaking everything. If you’re into DevOps culture, SRE practices, platform engineering, CI/CD, infrastructure automation, and how teams should think about reliability and security as things keep changing, this one should land. Guest Mazharul Islam (DevOps with Maz) UK-based DevOps Engineer. Posts a lot of hands-on content around cloud, DevOps fundamentals, and leveling up as an engineer. Links (Maz) YouTube: https://m.youtube.com/@devopswithmaz LinkedIn: https://www.linkedin.com/in/mazharul419 Topics we covered WHY behind DevOps, and why “tools” is the smallest part of it DevOps fundamentals vs tool-chasing Upskilling strategies for DevOps Engineers and SREs How to keep learning cloud and automation without drowning What strong teams measure and what “good” actually looks like (delivery, reliability, feedback loops) Agentic AI, AI agents in operations, and the next era of DevOps MCPs, automation guardrails, and safe ways to scale change How companies can “level up” their engineering org without turning it into chaos We also discussed the previous episode of Ship It Weekly - GitHub Runner Pricing Pause, Terraform Cloud Limits, and AI in CI https://www.tellerstech.com/ship-it-weekly/github-runner-pricing-pause-terraform-cloud-limits-and-ai-in-ci/ Book Maz recommended The DevOps Handbook: How to Create World-Class Agility, Reliability, and Security in Technology Organizations (Paperback, Oct 6, 2016) Gene Kim, Jez Humble, Patrick Debois, John Willis About Ship It Weekly (format) Ship It Weekly is for people running infrastructure and owning reliability. Most episodes are quick weekly news recaps for DevOps, SRE, and platform engineering. In between those weekly drops, I’ll publish interview episodes like this one. Subscribe / help the show If you want the weekly DevOps news recaps plus these interviews, hit follow or subscribe in your podcast app. And if you’re feeling generous, leave a rating or review and share this episode with a coworker (especially your on-call buddy). That stuff genuinely helps the show get discovered.

This week on Ship It Weekly, Brian looks at how the “platform tax” is showing up everywhere: pricing model shifts, CI dependencies, and new security boundaries thanks to AI agents. We start with GitHub Actions. GitHub announced a new “cloud platform” charge for self-hosted runners in private/internal repos… then hit pause after backlash. Hosted runner price reductions for 2026 are still planned. We also got the perfect timing joke: a GitHub incident the same week. Next up is HashiCorp. Legacy HCP Terraform (Terraform Cloud) Free is reaching end-of-life in 2026, with orgs moving to the newer Free tier capped at 500 managed resources. If you’re running real infrastructure, this is a good moment to audit what you’re actually managing and decide whether you’re cleaning up, paying, or planning a migration. Then we talk PromptPwnd: why stuffing untrusted PR/issue text into AI agent prompts (inside CI) can turn into a supply chain/security problem. The short version: treat AI inputs like hostile user input, keep tokens/permissions minimal, and don’t let agents “run with scissors.” We also cover the Home Depot report about long-lived access exposure as a reminder that secrets hygiene, blast radius, and detection still matter more than the shiny tools. In the lightning round: CDKTF is sunset/archived, Bitbucket is cleaning up free unused workspaces, and SourceHut is proposing pricing changes. We wrap with a human note on “platform whiplash” and why a simple watchlist beats carrying all this stuff in your head. Links from this episode GitHub Actions pricing + pause https://runs-on.com/blog/github-self-hosted-runner-fee-2026/ https://x.com/github/status/2001372894882918548 https://www.githubstatus.com/incidents/x696x0g4t85l HashiCorp / Terraform Cloud free plan changes https://github.com/hashicorp/terraform-cdk?tab=readme-ov-file#sunset-notice https://www.reddit.com/r/Terraform/s/slYm77wzYr PromptPwnd / AI agents in CI https://www.aikido.dev/blog/promptpwnd-github-actions-ai-agents Home Depot access exposure report https://techcrunch.com/2025/12/12/home-depot-exposed-access-to-internal-systems-for-a-year-says-researcher/ Bitbucket cleanup https://community.atlassian.com/forums/Bitbucket-articles/Bitbucket-cleanup-of-free-unused-workspaces-what-you-need-to/ba-p/3144063 SourceHut pricing proposal https://sourcehut.org/blog/2025-12-01-proposed-pricing-changes/

In this episode of Ship It Weekly, Brian powers through a cold and digs into a very “infra grown-up” week in DevOps. First up, IBM is buying Confluent for $11B. We talk about what that means if you’re on Confluent Cloud today, still running your own Kafka, or trying to choose between Confluent, MSK, and DIY. It’s part of a bigger pattern after IBM’s HashiCorp deal, and it has real implications for vendor concentration and “plan B” strategies. Then we shift to React2Shell, a 10.0 RCE in React Server Components that’s already being exploited in the wild. Even if you never touch React, if you run platforms or Kubernetes for teams using Next.js or RSC, you’re on the hook for patching windows, WAF rules, and blast-radius thinking. We also look at Netflix’s write-up on consolidating relational databases onto Aurora PostgreSQL, with big performance gains and cost savings. It’s a good excuse to step back and ask whether your own Postgres fleet still makes sense at the scale you’re at now. In the lightning round, we hit OpenTofu 1.11’s new language features, practical Terraform “tips from the trenches,” Ghostty becoming a non-profit project, and two spec-driven dev tools (Spec Kit and OpenSpec) that show what sane AI-assisted development might look like. For the human side, we close with “Your Brain on Incidents” and what high-stress outages actually do to people, plus a few concrete ideas for making on-call less brutal. If you’re on a platform team, own SLOs, or you’re the person people ping when “something is wrong with prod,” this one should give you a mix of immediate to-dos and longer-term questions for your roadmap. Links: IBM + Confluent https://www.confluent.io/blog/ibm-to-acquire-confluent/ https://newsroom.ibm.com/2025-12-08-ibm-to-acquire-confluent-to-create-smart-data-platform-for-enterprise-generative-ai React2Shell (CVE-2025-55182) https://react.dev/blog/2025/12/03/critical-security-vulnerability-in-react-server-components Netflix on Aurora PostgreSQL https://aws.amazon.com/blogs/database/netflix-consolidates-relational-database-infrastructure-on-amazon-aurora-achieving-up-to-75-improved-performance/ Tools & tips https://opentofu.org/blog/opentofu-1-11-0/ https://rosesecurity.dev/2025/12/04/terraform-tips-and-tricks.html https://mitchellh.com/writing/ghostty-non-profit https://github.com/github/spec-kit https://github.com/Fission-AI/OpenSpec Human side https://uptimelabs.io/your-brain-on-incidents/

In this episode of Ship It Weekly, Brian looks at re:Invent through a platform/SRE lens and pulls out the updates that actually change how you design and run systems. We talk about regional NAT Gateways and Route 53 Global Resolver on the networking side, ECS Express Mode and EKS Capabilities as new paved roads for app teams, S3 Vectors GA and 50 TB S3 objects for AI and data lakes, Aurora PostgreSQL dynamic data masking, CodeCommit’s return to full GA, and IAM Policy Autopilot for AI-assisted IAM policies. This was recorded mid–re:Invent, so consider it a “what matters so far” pass, not a full recap. Outside AWS, we get into Google’s 130,000-node GKE cluster and what actually applies if you’re running normal-sized clusters, plus the “It’s time to kill staging” argument and what responsible testing in production looks like with feature flags, progressive delivery, and solid observability. In the lightning round, we hit Zachary Loeber’s Terraform MCP server and terraform-ingest (letting AI tools speak your real Terraform modules), Runs-On’s EC2 instance rankings so you stop picking instance types by vibes, and Airbnb’s adaptive traffic management for their key-value store. We close with Nolan Lawson’s “The fate of small open source” and what it means when your platform quietly depends on one-maintainer libraries. Links from this episode: AWS highlights: https://aws.amazon.com/about-aws/whats-new/2025/11/aws-nat-gateway-regional-availability https://aws.amazon.com/blogs/aws/introducing-amazon-route-53-global-resolver-for-secure-anycast-dns-resolution-preview https://aws.amazon.com/about-aws/whats-new/2025/11/announcing-amazon-ecs-express-mode https://aws.amazon.com/about-aws/whats-new/2025/12/amazon-s3-vectors-generally-available/ Other topics: https://cloud.google.com/blog/products/containers-kubernetes/how-we-built-a-130000-node-gke-cluster https://thenewstack.io/its-time-to-kill-staging-the-case-for-testing-in-production/ https://blog.zacharyloeber.com/article/terraform-custom-module-mcp-server/ https://go.runs-on.com/instances/ranking https://medium.com/airbnb-engineering/from-static-rate-limiting-to-adaptive-traffic-management-in-airbnbs-key-value-store-29362764e5c2 https://nolanlawson.com/2025/11/16/the-fate-of-small-open-source/