|
| 1 | +--- |
| 2 | +title: March 2026 |
| 3 | +weight: -202603 |
| 4 | +outline: 2 |
| 5 | +--- |
| 6 | + |
| 7 | +# Hack The Garden Sofia Edition 03/2026 Wrap Up |
| 8 | + |
| 9 | +- 🗓️ **Date:** 16.03.2026 – 20.03.2026 |
| 10 | +- 📍 **Location:** [SAP Center Sofia](https://maps.app.goo.gl/SPdvQ4F2p7Qqfx4p9) |
| 11 | +- 👤 **Organizer:** [SAP](https://www.sap.com/) |
| 12 | +- 📘 **Topics:** [hackathon/discussions#41](https://github.com/gardener/hackathon/discussions/41) |
| 13 | + |
| 14 | + |
| 15 | + |
| 16 | +## 🐳 [GEP-28] Self-Hosted Shoot: Gardener-in-Docker (`gind`) |
| 17 | + |
| 18 | +> [!TIP] |
| 19 | +> You can find out more about [Self-Hosted Shoot Clusters in GEP-28](https://github.com/gardener/enhancements/blob/main/geps/0028-self-hosted-shoot-clusters). |
| 20 | +
|
| 21 | +> **Tracking:** [hackathon#8](https://github.com/gardener/hackathon/issues/8) |
| 22 | +
|
| 23 | +### Problem Statement |
| 24 | + |
| 25 | +It should be possible to create self-hosted shoot clusters using `gardenadm` and run Gardener inside such a cluster. |
| 26 | +Before introducing a tool like `gind` (which runs the self-hosted shoot directly in Docker), we first need to support hosting Gardener inside a self-hosted shoot cluster. |
| 27 | + |
| 28 | +### Achievements |
| 29 | + |
| 30 | +* Deployed `gardener-operator` into the self-hosted shoot. |
| 31 | +* Deployed a `Garden` resource — the self-hosted shoot now serves as runtime cluster for the virtual garden. |
| 32 | +* Enabled the `ManagedSeed` controller in the shoot `gardenlet`, allowing the self-hosted `Shoot` itself to be referenced in the `ManagedSeed`. |
| 33 | +* Adapted the local setup for direct API access to both the self-hosted shoot API server and the virtual garden API server from the host machine (no port-forwarding). |
| 34 | + |
| 35 | +### Next Steps |
| 36 | + |
| 37 | +* Cleanup the code and commits, adapt documentation and `make` rules. |
| 38 | +* Open individual PRs for the different features and get them merged. |
| 39 | +* Introduce an e2e test for this scenario ("fully self-contained Gardener"). |
| 40 | +* Try spinning up workerless and regular hosted shoots on this seed. |
| 41 | + |
| 42 | +### Code & Pull Requests |
| 43 | + |
| 44 | +* [Don't create duplicate `ControllerInstallation`s when self-hosted `Shoot` is also a `Seed` – gardener/gardener#14282](https://github.com/gardener/gardener/pull/14282) |
| 45 | +* [Ensure all system pods run on system component nodes – gardener/gardener#14367](https://github.com/gardener/gardener/pull/14367) |
| 46 | +* [`gardenadm connect`: Enable `ManagedSeed` controller in `gardenlet` – gardener/gardener#14369](https://github.com/gardener/gardener/pull/14369) |
| 47 | +* [Make self-hosted shoot API server accessible from host machine - gardener/gardener#14370](https://github.com/gardener/gardener/pull/14370) |
| 48 | +* [Enable deployment of `gardener-operator` (and `Garden`) inside self-hosted `Shoot`s – gardener/gardener#14387](https://github.com/gardener/gardener/pull/14387) |
| 49 | +* [Introduce GinD (Gardener-in-Docker) dev setup for self-hosted `Shoot`s w/ unmanaged infra – gardener/gardener#14700](https://github.com/gardener/gardener/issues/14700) |
| 50 | +* [Create `ManagedSeed` for self-hosted `Shoot` to promote it to seed cluster – gardener/gardener#14747](https://github.com/gardener/gardener/pull/14747) |
| 51 | + |
| 52 | +## 🕹️ [GEP-28] Ensure System Pods Run on Control Plane Nodes |
| 53 | + |
| 54 | +> **Tracking:** [hackathon#16](https://github.com/gardener/hackathon/issues/16) |
| 55 | +
|
| 56 | +### Problem Statement |
| 57 | + |
| 58 | +System components in a self-hosted shoot (pods in `garden` namespace, extensions, system pods in `kube-system`) are not guaranteed to run exclusively on control plane nodes. |
| 59 | +Over time, they might get rescheduled to worker nodes. |
| 60 | + |
| 61 | +### Achievements |
| 62 | + |
| 63 | +* Implemented placement enforcement so that system pods exclusively run on control plane nodes. |
| 64 | + |
| 65 | +### Code & Pull Requests |
| 66 | + |
| 67 | +* [Ensure all system pods run on system component nodes – gardener/gardener#14367](https://github.com/gardener/gardener/pull/14367) |
| 68 | + |
| 69 | +## ❤️🩹 [GEP-28] Self-Hosted Shoot Control Plane Restoration |
| 70 | + |
| 71 | +> **Tracking:** [hackathon#22](https://github.com/gardener/hackathon/issues/22) |
| 72 | +
|
| 73 | +### Problem Statement |
| 74 | + |
| 75 | +When a self-hosted shoot cluster loses its control plane, it must be possible to restore the secrets and control plane state from the `ShootState` resource. |
| 76 | + |
| 77 | +### Achievements |
| 78 | + |
| 79 | +* PoC branch using the `ShootState` for restoring secrets when restoring a control plane of a self-hosted shoot: [poc/gep-28-dr](https://github.com/ialidzhikov/gardener/commits/poc/gep-28-dr/). |
| 80 | +* Demo scripts: [demo/restore-from-shootstate](https://github.com/ialidzhikov/gardener/commits/demo/restore-from-shootstate/). |
| 81 | +* Fixed a bug when computing the ShootState for a self-hosted shoot cluster. |
| 82 | + |
| 83 | +### Next Steps |
| 84 | + |
| 85 | +* Eliminate hacks/workarounds: enable etcd encryption, adapt csrapprover for gardener-node-agent CSRs, fix pod network availability check, sanitize etcd data, eliminate second-phase restore. |
| 86 | +* Design and implement how to read/compute the ShootState (via `gardenadm discover` or from etcd backup). |
| 87 | +* Design and implement etcd backup restore. |
| 88 | +* Add support for restoring a self-hosted shoot with worker nodes. |
| 89 | + |
| 90 | +### Code & Pull Requests |
| 91 | + |
| 92 | +* [Fix ShootState resource deployment for a self-hosted Shoot – gardener/gardener#14339](https://github.com/gardener/gardener/pull/14339) |
| 93 | + |
| 94 | +## 🗝️ [GEP-28] Eliminate Static Admin Token After `gardenadm connect` |
| 95 | + |
| 96 | +> **Tracking:** [hackathon#14](https://github.com/gardener/hackathon/issues/14) |
| 97 | +
|
| 98 | +### Problem Statement |
| 99 | + |
| 100 | +After `gardenadm init`, the control plane components use a static token with cluster-admin privileges for bootstrapping. |
| 101 | +Once the cluster is fully connected (`gardenadm connect`), this should be replaced with short-lived tokens from `gardener-resource-manager`. |
| 102 | + |
| 103 | +### Outcome |
| 104 | + |
| 105 | +Discussed and decided that this will be part of the `shoot/shoot` controller and should not be handled in `gardenadm init` or `gardenadm connect` explicitly. |
| 106 | +Closed in favor of [Experiment with `shoot/shoot` controller in Self-Hosted Shoot Clusters – hackathon#45](https://github.com/gardener/hackathon/issues/45). |
| 107 | + |
| 108 | +## 🔑 Functional Local Setup with Workload Identity |
| 109 | + |
| 110 | +> **Tracking:** [hackathon#28](https://github.com/gardener/hackathon/issues/28) |
| 111 | +
|
| 112 | +### Problem Statement |
| 113 | + |
| 114 | +The local development setup does not work with Workload Identity (WI), making it impossible to test WI-dependent scenarios locally. |
| 115 | + |
| 116 | +### Achievements |
| 117 | + |
| 118 | +* Initial implementation establishing trust between the local KinD cluster and the Gardener Workload Identity Issuer. |
| 119 | +* Identified that Machine Controller Manager is not deployed with minimal permissions — opened a PR to address this. |
| 120 | + |
| 121 | +### Next Steps |
| 122 | + |
| 123 | +* Clean up the code in the linked branch. |
| 124 | +* Verify all scenarios work as expected and current tests pass. |
| 125 | +* Implement new e2e tests leveraging Workload Identity or enable it for existing ones. |
| 126 | +* Add support for Workload Identity in other local scenarios (ETCD backups, DNS, etc.). |
| 127 | + |
| 128 | +### Code & Pull Requests |
| 129 | + |
| 130 | +* Branch: [dimityrmirchev/gardener:wi-local-setup](https://github.com/dimityrmirchev/gardener/tree/wi-local-setup) |
| 131 | +* [Reduce MCM permissions – gardener/gardener#14372](https://github.com/gardener/gardener/pull/14372) |
| 132 | + |
| 133 | +## 🤖 AGENTS.md / SKILLS.md for Gardener Repos |
| 134 | + |
| 135 | +> **Tracking:** [hackathon#31](https://github.com/gardener/hackathon/issues/31) |
| 136 | +
|
| 137 | +### Problem Statement |
| 138 | + |
| 139 | +AI-native development tools (Claude Code, Codex CLI, Gemini CLI) benefit from repository-level context files (`AGENTS.md`, `SKILLS.md`). |
| 140 | +The question is how to best leverage these for the Gardener ecosystem. |
| 141 | + |
| 142 | +### Achievements |
| 143 | + |
| 144 | +* Researched recent papers: *Evaluating AGENTS.md* (Feb 2026) and *SkillsBench* (Feb/Mar 2026). |
| 145 | +* Key findings: curated skills provide +16.2pp average improvement; LLM-generated context provides negligible or negative benefit; focused skills with 2–3 modules outperform comprehensive documentation. |
| 146 | +* Proposed a minimal `AGENTS.md` template focused on "common mistakes and confusion points" rather than comprehensive documentation. |
| 147 | + |
| 148 | +### Next Steps |
| 149 | + |
| 150 | +* Experiment with proposed `AGENTS.md` file in gardener org repos (not `gardener/gardener`). |
| 151 | +* If significant benefit is observed, present findings in a larger forum (Gardener Review Meeting). |
| 152 | + |
| 153 | +## 📦 PoC: Repo Tools Integration with Extension Repositories |
| 154 | + |
| 155 | +> **Tracking:** [hackathon#18](https://github.com/gardener/hackathon/issues/18) |
| 156 | +
|
| 157 | +### Problem Statement |
| 158 | + |
| 159 | +Extension repositories share ~10 almost identical make targets and hack scripts with `gardener/gardener`. |
| 160 | +Changes to these shared scripts result in copy-paste effort across all repositories (~20 PRs for a single fix). |
| 161 | + |
| 162 | +### Achievements |
| 163 | + |
| 164 | +* Explored a subtree approach for centralizing shared make targets and hack scripts into a separate repository. |
| 165 | +* Adapted `gardener-extension-shoot-rsyslog-relp` and `pvc-autoscaler` as PoC repositories. |
| 166 | + |
| 167 | +### Next Steps |
| 168 | + |
| 169 | +* Adapt additional repositories to validate the approach and catch problems early. |
| 170 | +* Gather feature requests based on newfound use cases. |
| 171 | + |
| 172 | +## ✅ Diki as a Service |
| 173 | + |
| 174 | +> **Tracking:** [hackathon#24](https://github.com/gardener/hackathon/issues/24) |
| 175 | +
|
| 176 | +### Problem Statement |
| 177 | + |
| 178 | +[Diki](https://github.com/gardener/diki) compliance checks should be schedulable, exportable, and operable as a service rather than one-off CLI runs. |
| 179 | + |
| 180 | +### Achievements |
| 181 | + |
| 182 | +* Merged previous work (first PoC and `diki-exporter`) into a working operator: [hackathon-poc branch](https://github.com/georgibaltiev/diki-operator/tree/hackathon-poc). |
| 183 | +* Implemented Postgres exporter in `diki-exporter`. |
| 184 | +* Made the operator capable of running in a different cluster than the `ComplianceScan`s (needed for seed/shoot-namespace topology). |
| 185 | +* PoC'ed `ScheduledComplianceScan`s — spawns scans based on a cron expression. |
| 186 | + |
| 187 | +### Next Steps |
| 188 | + |
| 189 | +* More testing and cleanup. |
| 190 | +* Pour the work into a Gardener extension. |
| 191 | + |
| 192 | +## 🧩 Extension: Generic Shoot Pack (CloudNativePG et al.) |
| 193 | + |
| 194 | +> **Tracking:** [hackathon#19](https://github.com/gardener/hackathon/issues/19) |
| 195 | +
|
| 196 | +### Problem Statement |
| 197 | + |
| 198 | +Installing upstream operators into shoot clusters requires repetitive per-operator extension development. |
| 199 | +A generic packaging mechanism would reduce this overhead. |
| 200 | + |
| 201 | +### Achievements |
| 202 | + |
| 203 | +* Developed [gardener-extension-shoot-pack](https://github.com/dnaeon/gardener-extension-shoot-pack/tree/feat/initial) — a generic Gardener extension that uses package specifications to install operators as managed resources. |
| 204 | +* Ships packages for: cert-manager, CloudNativePG, Prometheus Operator, and Valkey Operator. |
| 205 | +* Tooling available to inspect, view, and create new package specs. |
| 206 | + |
| 207 | +### Next Steps |
| 208 | + |
| 209 | +* Clean things up. |
| 210 | +* Add more tests. |
| 211 | + |
| 212 | +## 🪣 Fix Leaking ValidatingWebhookConfigurations in (Virtual-)Garden |
| 213 | + |
| 214 | +> **Tracking:** [hackathon#21](https://github.com/gardener/hackathon/issues/21) |
| 215 | +
|
| 216 | +### Problem Statement |
| 217 | + |
| 218 | +When deploying extensions via `gardener-operator` using `extensions.operator.gardener.cloud` resources, `ValidatingWebhookConfiguration`s remain in the virtual-garden cluster even after removing the extension. |
| 219 | +The root cause: the `--webhook-config-owner-namespace` option defaults to `garden` namespace, preventing proper garbage collection. |
| 220 | + |
| 221 | +### Achievements |
| 222 | + |
| 223 | +* Identified the root cause — missing `--webhook-config-owner-namespace` flag in extension admission deployments. |
| 224 | +* A fix existed since ~2 years ([Cleanup webhook configuration from virtual cluster when removing admission deployment – gardener/gardener#10585](https://github.com/gardener/gardener/pull/10585)) but was not adopted by all extensions. |
| 225 | +* Opened umbrella issue to track fixes across all affected extensions. |
| 226 | + |
| 227 | +### Next Steps |
| 228 | + |
| 229 | +* Apply the `--webhook-config-owner-namespace` option to each affected extension's admission deployment. |
| 230 | + |
| 231 | +### Code & Pull Requests |
| 232 | + |
| 233 | +* [Fix leaking of `validatingwebhookconfiguration` resources in extensions – gardener/gardener#14334](https://github.com/gardener/gardener/issues/14334) |
| 234 | + |
| 235 | +## 📡 Resolve the Istio Metrics Leak |
| 236 | + |
| 237 | +> **Tracking:** [hackathon#12](https://github.com/gardener/hackathon/issues/12) |
| 238 | +
|
| 239 | +### Problem Statement |
| 240 | + |
| 241 | +Istio sidecar metrics for terminated pods accumulate indefinitely, leading to unbounded cardinality in Prometheus. |
| 242 | + |
| 243 | +### Achievements |
| 244 | + |
| 245 | +* Configured Istio metric rotation via environment variables (`METRIC_ROTATION_INTERVAL`, `METRIC_GRACEFUL_DELETION_INTERVAL`). |
| 246 | +* Verified correct behavior: after rotation interval, old pod metrics disappear and new pod metrics appear; long-lived connection metrics reset correctly (counters restart from 0, compatible with PromQL `rate` functions). |
| 247 | +* Fixed duplicate scraping caused by two Istio services (`istio-gateway` LoadBalancer and `istio-gateway-internal` ClusterIP) matching the same ServiceMonitor label selector. |
| 248 | + |
| 249 | +### Next Steps |
| 250 | + |
| 251 | +* The env-var approach is deprecated from Istio v1.28+. |
| 252 | +* Migration to annotation-based configuration (`SidecarStatsEvictionInterval`) will be needed when upgrading beyond v1.27. |
| 253 | + |
| 254 | +### Code & Pull Requests |
| 255 | + |
| 256 | +* [Evict stale Istio Gateway metrics and reenable scraping – gardener/gardener#14337](https://github.com/gardener/gardener/pull/14337) |
| 257 | + |
| 258 | +--- |
| 259 | + |
| 260 | + |
| 261 | + |
| 262 | +--- |
| 263 | + |
| 264 | + |
0 commit comments