2 Commits

Author	SHA1	Message	Date
jgrusewski	33b4020e06	feat(monitoring): add cockpit dashboard, migrate all dashboards to dynamic datasources ... Add foxhunt-cockpit.json — single-page overview with 7 rows (health bar, service status, resources, GPU/training, restarts/network, errors, live logs) pulling critical signals from all domain dashboards. Migrate all 10 dashboards from hardcoded datasource UIDs to ${DS_PROMETHEUS}, ${DS_LOKI}, ${DS_TEMPO} variables with __inputs sections for portable import. Update import.sh to auto-detect datasource UIDs from Grafana API at deploy time, using /api/dashboards/import endpoint. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-28 23:04:42 +01:00
jgrusewski	4c54fe0e37	feat(infra): add Cockpit Grafana dashboards for K8s, GPU, and GitLab ... - cluster-overview: nodes, CPU, memory, disk, network per node/namespace - gpu-overview: DCGM utilization, VRAM, temperature, power, per-pod - gitlab-services: puma, sidekiq, gitaly + foxhunt service resource usage - import.sh: one-shot script to push all dashboards via Grafana API Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-27 01:22:24 +01:00