tegwick
2c2a6073ff
T06: optimization-case schema, fail-closed evaluator, cadence and decision template. Every option including the baseline must present all ten decision fields; one unknown blocks the comparison. Validated on the storage case (Hetzner computes and loses to Scaleway by EUR 29.14/month on operator labour; Host Europe blocks on four named gaps) and on the non-storage reef-railiance k3s rightsizing case (low utilization is real, but nothing is costable while the railiance01 price is unknown). T07: portfolio report over coverage, lifecycle, utilization, cost, renewals, risks, open cases, and next actions, derived only from committed evidence. Portfolio spend is reported null rather than as a partial sum, unattributed cost is a named list rather than a spread, and unmeasurable resources are reported rather than dropped. RESOURCE-WP-0003 is finished; both cases remain blocked_on_evidence against live delegated records in other repositories. RESOURCE-WP-0002 is untouched. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
200 lines
7.7 KiB
JSON
200 lines
7.7 KiB
JSON
{
|
|
"schema_version": "0.1",
|
|
"record_scope": "operational",
|
|
"case_id": "opt:reef-railiance-k3s:2026-08",
|
|
"case_type": "rightsizing",
|
|
"trigger": "cadence",
|
|
"review_period": "2026-08",
|
|
"created_at": "2026-08-14T00:00:00Z",
|
|
"resource_ids": [
|
|
"resource:railiance:reef-railiance:k3s",
|
|
"resource:hosteurope:railiance01"
|
|
],
|
|
"baseline": {
|
|
"option_id": "railiance01-single-node",
|
|
"label": "Current single-node k3s on the Host Europe railiance01 server (4 vCPU, 15.62 GiB)",
|
|
"one_time_eur": 0,
|
|
"recurring_infrastructure_eur_month": null,
|
|
"recurring_internal_labor_eur_month": null,
|
|
"recurring_external_labor_eur_month": 0,
|
|
"utilization": {
|
|
"cpu": {
|
|
"provisioned": { "value": 4, "unit": "vCPU" },
|
|
"used": { "value": 0.564, "unit": "vCPU" }
|
|
},
|
|
"memory": {
|
|
"provisioned": { "value": 15.62, "unit": "GiB" },
|
|
"used": { "value": 5.83, "unit": "GiB" }
|
|
},
|
|
"root_filesystem": {
|
|
"provisioned": { "value": 192.69, "unit": "GiB" },
|
|
"used": { "value": 66.77, "unit": "GiB" }
|
|
}
|
|
},
|
|
"uncertainty": {
|
|
"level": "high",
|
|
"notes": [
|
|
"Capacity and usage are a single live observation on 2026-08-11, not a utilization history",
|
|
"A one-off sample cannot distinguish steady-state headroom from a quiet moment; peak and growth are unknown",
|
|
"The server carries Helix Forge, Coulomb Social, rapp-qonto, and shared platform services, so downsizing headroom is shared, not per-workload"
|
|
]
|
|
},
|
|
"service_constraints": {
|
|
"nodes": { "value": 1, "unit": "count" },
|
|
"pod_limit": { "value": 110, "unit": "count" }
|
|
},
|
|
"failure_domains": [
|
|
"provider:host-europe",
|
|
"host:railiance01",
|
|
"cluster:reef-railiance"
|
|
],
|
|
"exit_path": "Provision replacement capacity, restore or redeploy the cluster and stateful services, switch DNS and ingress, verify workloads, then cancel and erase the server.",
|
|
"unknowns": [
|
|
"railiance01 product, booked monthly price, tax treatment, and renewal or cancellation dates (owner: railiance-infra, RAIL-HO-WP-0008)",
|
|
"cluster operations labour hours per month (owner: railiance-cluster, RAIL-BS-WP-0014)",
|
|
"requested-versus-used capacity history and peak load (owner: railiance-cluster, RAIL-BS-WP-0014)",
|
|
"declared service objectives for the workloads sharing the node (owner: workload repositories)"
|
|
]
|
|
},
|
|
"alternatives": [
|
|
{
|
|
"option_id": "smaller-single-server",
|
|
"label": "Consolidate onto a smaller Host Europe server sized to observed usage plus headroom",
|
|
"one_time_eur": null,
|
|
"recurring_infrastructure_eur_month": null,
|
|
"recurring_internal_labor_eur_month": null,
|
|
"recurring_external_labor_eur_month": 0,
|
|
"utilization": {
|
|
"cpu": {
|
|
"provisioned": { "value": null, "unit": "vCPU" },
|
|
"used": { "value": 0.564, "unit": "vCPU" }
|
|
},
|
|
"memory": {
|
|
"provisioned": { "value": null, "unit": "GiB" },
|
|
"used": { "value": 5.83, "unit": "GiB" }
|
|
}
|
|
},
|
|
"uncertainty": {
|
|
"level": "high",
|
|
"notes": [
|
|
"No target instance can be sized without peak-load history; observed mean usage is not a sizing basis",
|
|
"Migration requires a full cluster rebuild and stateful restore, so one-time labour is material and unmeasured",
|
|
"Retains the single-host failure domain the baseline already carries"
|
|
]
|
|
},
|
|
"service_constraints": {
|
|
"nodes": { "value": 1, "unit": "count" }
|
|
},
|
|
"failure_domains": [
|
|
"provider:host-europe",
|
|
"host:railiance01",
|
|
"cluster:reef-railiance"
|
|
],
|
|
"exit_path": null,
|
|
"unknowns": [
|
|
"Host Europe server catalogue, sizes, and prices available to this account (owner: railiance-infra, RAIL-HO-WP-0008)",
|
|
"peak CPU and memory over a full billing period, required to size the target (owner: railiance-cluster, RAIL-BS-WP-0014)",
|
|
"one-time migration labour and workload downtime budget (owner: railiance-cluster)"
|
|
]
|
|
},
|
|
{
|
|
"option_id": "threephoenix-ha-cluster",
|
|
"label": "Three-node HA cluster per RAIL-BS-WP-0007, removing the single-host failure domain",
|
|
"one_time_eur": null,
|
|
"recurring_infrastructure_eur_month": null,
|
|
"recurring_internal_labor_eur_month": null,
|
|
"recurring_external_labor_eur_month": 0,
|
|
"utilization": {
|
|
"cpu": {
|
|
"provisioned": { "value": null, "unit": "vCPU" },
|
|
"used": { "value": 0.564, "unit": "vCPU" }
|
|
},
|
|
"memory": {
|
|
"provisioned": { "value": null, "unit": "GiB" },
|
|
"used": { "value": 5.83, "unit": "GiB" }
|
|
}
|
|
},
|
|
"uncertainty": {
|
|
"level": "high",
|
|
"notes": [
|
|
"This option raises recurring cost by design; it is justified by availability, not by saving, and payback is the wrong test for it",
|
|
"Node count and placement are owned by RAIL-BS-WP-0007, not by this case"
|
|
]
|
|
},
|
|
"service_constraints": {
|
|
"nodes": { "value": 3, "unit": "count" }
|
|
},
|
|
"failure_domains": [
|
|
"provider:host-europe",
|
|
"cluster:reef-railiance"
|
|
],
|
|
"exit_path": null,
|
|
"unknowns": [
|
|
"per-node infrastructure price and whether nodes span independent failure domains (owner: railiance-cluster, RAIL-BS-WP-0014)",
|
|
"operations labour for a multi-node cluster versus a single node (owner: railiance-cluster)",
|
|
"declared availability objective that would justify the added recurring cost (owner: workload repositories)",
|
|
"replicated-storage requirement created by removing local-path storage (owner: railiance-cluster)"
|
|
]
|
|
}
|
|
],
|
|
"decision": {
|
|
"state": "blocked_on_evidence",
|
|
"recommended_option_id": null,
|
|
"rationale": "Observed utilization is genuinely low: 14 percent of CPU, 37 percent of memory, and 35 percent of the root filesystem on a single sample. That is a real rightsizing signal, and it is not sufficient to act on. No option can be costed because the railiance01 booked price is unknown, so no saving, no payback, and no comparison between downsizing and the HA alternative can be computed. The case stays open and blocked against the named delegated workplans rather than producing a rightsizing recommendation from a mean-usage sample.",
|
|
"approver": null,
|
|
"approved_on": null,
|
|
"delegated_to": [
|
|
"railiance-infra",
|
|
"railiance-cluster"
|
|
]
|
|
},
|
|
"financial_handoff": {
|
|
"cost_attribution_key": null,
|
|
"sent": false,
|
|
"reference": null
|
|
},
|
|
"outcome": {
|
|
"feeds_forecast": [
|
|
"examples/control-cycle/cluster-forecast.json"
|
|
],
|
|
"actual_refs": []
|
|
},
|
|
"evidence": [
|
|
{
|
|
"kind": "telemetry",
|
|
"ref": "kubernetes:reef-railiance/node/239.62.205.92.host.secureserver.net@2026-08-11T09:05:25Z",
|
|
"authority": "reef-railiance Kubernetes API",
|
|
"observed_at": "2026-08-11"
|
|
},
|
|
{
|
|
"kind": "telemetry",
|
|
"ref": "ssh:railiance01 host capacity observation 2026-08-11T09:05:25Z",
|
|
"authority": "railiance01 operating system",
|
|
"observed_at": "2026-08-11"
|
|
},
|
|
{
|
|
"kind": "document",
|
|
"ref": "docs/evidence/RESOURCE-WP-0003-initial-portfolio-discovery-2026-08-11.md",
|
|
"authority": "resource-control",
|
|
"observed_at": "2026-08-11"
|
|
},
|
|
{
|
|
"kind": "workplan",
|
|
"ref": "RAIL-HO-WP-0008",
|
|
"authority": "railiance-infra",
|
|
"observed_at": null
|
|
},
|
|
{
|
|
"kind": "workplan",
|
|
"ref": "RAIL-BS-WP-0014",
|
|
"authority": "railiance-cluster",
|
|
"observed_at": null
|
|
},
|
|
{
|
|
"kind": "workplan",
|
|
"ref": "RAIL-BS-WP-0007",
|
|
"authority": "railiance-cluster",
|
|
"observed_at": null
|
|
}
|
|
]
|
|
}
|