Graham McIntire 0c5cbbdc54 Bump prop pod memory limit 4Gi → 6Gi ... Even with the narrow HRRR grid pressure levels (4fa6798), pods are still OOMKilled partway through the propagation chain. After the HRRR fetch, the grid worker pipes 92k grid points through native duct metric merge (~100 MB held for ~2 minutes), builds the full weather grid cache (92k rows × ~40 fields), runs compute_scores (460k score maps across 5 bands materialized in a list), and writes five .ntms files — all while ERA5Submit workers keep churning through CDS backfill in background processes. Nodes have ~42% memory headroom on limits; the 2 Gi bump fits comfortably. Shrinking the in-process allocations is a better long-term fix but needs a proper pass through compute_scores and persist_profiles. This gets prod unstuck in the meantime so the forecast timeline can actually populate.		2026-04-15 08:29:09 -05:00
..
deployment.yaml	Bump prop pod memory limit 4Gi → 6Gi	2026-04-15 08:29:09 -05:00
flux.yaml	Add k8s manifests and Forgejo CI for cluster deployment	2026-04-06 09:25:24 -05:00
kustomization.yaml	Drop secret.yaml from kustomization to unblock Flux reconciliation	2026-04-12 15:05:48 -05:00
namespace.yaml	Add k8s manifests and Forgejo CI for cluster deployment	2026-04-06 09:25:24 -05:00
rbac.yaml	Add Erlang clustering via libcluster and expose LiveDashboard	2026-04-06 09:46:48 -05:00
service.yaml	Add session affinity to service for stable LiveView websockets	2026-04-06 09:48:24 -05:00