Review provisional Vergabe sizing against current node budget

Assistant: codex
Assistant-Model: gpt-6-astra
Assistant-Session: 01a15374-f59f-7a31-86d2-d23b5d2ed181
This commit is contained in:
tegwick 2026-10-19 15:07:20 +02:00
parent 9c57b0d96b
commit 565b16e2d0
4 changed files with 118 additions and 1 deletions

View file

@ -15,6 +15,7 @@
| workplan | RESOURCE-WP-0005 | finished | — | workplans/RESOURCE-WP-0005-resource-procurement-facility.md |
| workplan | RESOURCE-WP-0006 | finished | — | workplans/RESOURCE-WP-0006-resource-object-and-reef-refs.md |
| workplan | RESOURCE-WP-0007 | finished | — | workplans/RESOURCE-WP-0007-state-hub-release-margin.md |
| workplan | RESOURCE-WP-0008 | active | — | workplans/RESOURCE-WP-0008-custodian-allocation-review.md |
| task | RESOURCE-WP-0001-T01 | done | — | workplans/RESOURCE-WP-0001-statehub-bootstrap.md |
| task | RESOURCE-WP-0001-T02 | done | — | workplans/RESOURCE-WP-0001-statehub-bootstrap.md |
| task | RESOURCE-WP-0001-T03 | done | — | workplans/RESOURCE-WP-0001-statehub-bootstrap.md |
@ -49,3 +50,6 @@
| task | RESOURCE-WP-0006-T03 | done | — | workplans/RESOURCE-WP-0006-resource-object-and-reef-refs.md |
| task | RESOURCE-WP-0006-T04 | done | — | workplans/RESOURCE-WP-0006-resource-object-and-reef-refs.md |
| task | RESOURCE-WP-0007-T01 | done | — | workplans/RESOURCE-WP-0007-state-hub-release-margin.md |
| task | RESOURCE-WP-0008-T01 | done | — | workplans/RESOURCE-WP-0008-custodian-allocation-review.md |
| task | RESOURCE-WP-0008-T02 | progress | — | workplans/RESOURCE-WP-0008-custodian-allocation-review.md |
| task | RESOURCE-WP-0008-T03 | todo | — | workplans/RESOURCE-WP-0008-custodian-allocation-review.md |

View file

@ -57,6 +57,7 @@
"evidence": [
{"kind": "telemetry", "ref": "kubernetes:reef-railiance/node/239.62.205.92.host.secureserver.net@2026-08-11T09:05:25Z", "observed_at": "2026-08-11", "authority": "reef-railiance Kubernetes API"},
{"kind": "workload", "ref": "/home/worsch/railiance-cluster/docs/rail-kubernetes-substrate-profile.md", "observed_at": "2026-08-11", "authority": "railiance-cluster"},
{"kind": "workload", "ref": "STATE-WP-0091: state-hub MCP CPU request 50m→10m after partial telemetry (p99 2.35m, peak 30.45m, 1078 samples). 105m node remaining is not reef admission and does not authorize procurement or scaling.", "observed_at": "2026-09-14", "authority": "state-hub"}
{"kind": "workload", "ref": "STATE-WP-0091: state-hub MCP CPU request 50m→10m after partial telemetry (p99 2.35m, peak 30.45m, 1078 samples). 105m node remaining is not reef admission and does not authorize procurement or scaling.", "observed_at": "2026-09-14", "authority": "state-hub"},
{"kind": "workload", "ref": "docs/evidence/2026-10-19-custodian-allocation-review.md", "observed_at": "2026-10-19", "authority": "resource-control review of Custodian8e67cfb and Vergabe11f7c8a; reservation snapshot and provisional synthetic evidence, not admission"}
]
}

View file

@ -0,0 +1,41 @@
# Initial allocation review — 2026-10-19
Provider: resource-control; beneficiary CUST-WP-0071-T03; receiving record
RESOURCE-WP-0008. This is an initial review, not an approved resize.
Sources: Custodian `8e67cfb`, `docs/evidence/2026-10-19-allocation-reconcile.json`
and its provenance; Vergabe `11f7c8a`,
`docs/evidence/2026-10-19-synthetic-sizing/`. Kubernetes target identity was
verified before the read-only capture at 12:50:41Z. A separate node description
agrees with 3605m scheduled CPU requests and 9856Mi memory requests.
| Reservation scenario | Scheduled CPU | Residual against 4000m |
| --- | ---: | ---: |
| Current snapshot, including Vergabe60m | 3605m | 395m |
| Replace Vergabe60m with experimental1950m | 5495m | -1495m |
The second row is an infeasible reservation scenario, not a proposal to raise
capacity or limits. Host and Kubernetes CPUs are the same four vCPUs and are
counted once. The 395m remainder is not guaranteed execution or rollout room.
STATE-WP-0091's passing preflight models its own release sequence only.
The synthetic run used two users/two workers,100 initial tenders,1MiB files,
0.2-second cycle think time,30s idle and180.889s work on an unconstrained local
Ryzen3500U host. It completed3028 operations without error, ordinary p95 227ms.
Aggregate app CPU p95 was1298m and sampled peak1917m. Tunable margins derive the
1950m request/2900m limit envelope; this is hardware/workload-specific and exceeds
the current production1000m CPU limit. Peak summed RSS194MiB includes shared-page
double counting; the calculated256Mi request/320Mi limit does not justify lowering
the existing1Gi limit without startup/restart and target-runtime evidence.
Current disposition: do not apply the unconstrained envelope. Retain the existing
explicitly accepted prototype settings pending the isolated1000m/1Gi comparison
and owner review. Passing a quota test would not establish that60m request protects
latency under production contention. Database demand remains a separate consumer.
No financial attribution, procurement or new spending decision follows.
Remaining budget inputs: thirteen grouped zero-request workloads; unknown namespace
owners for mas/playwright/rapp-qonto; representative shared-service and build-runner
demand; host/rollout margin; current contention and throttle evidence. Historical
portfolio utilization remains dated2026-08-11 and is not silently refreshed by this
reservation snapshot. RESOURCE-WP-0008 keeps the allocation review active.

View file

@ -0,0 +1,71 @@
---
id: RESOURCE-WP-0008
type: workplan
title: "Review configurable Vergabe sizing against the Railiance allocation budget"
domain: financials
repo: resource-control
status: active
owner: codex
topic_slug: railiance
created: "2026-10-19"
updated: "2026-10-19"
flavor: implementation
related: [CUST-WP-0071, VERGABE-WP-0019, RAPPS-WP-0014, RCLUSTER-WP-0014]
depends_on: []
---
Provider delivery for CUST-WP-0071-T03. The user authorized direct owner work and
configurable synthetic provisional sizing on 2026-10-19. Reuse the existing
portfolio resource and telemetry surfaces. This is current demanded work, not
promotion of an old residual. No procurement, capacity admission or resource
mutation is authorized by this record.
## Reconcile the measured inputs and current reservation budget
```task
id: RESOURCE-WP-0008-T01
status: done
priority: high
```
Retained `docs/evidence/2026-10-19-custodian-allocation-review.md` links the exact
owner evidence and explains the current 395m reservation remainder. The first
unbounded local synthetic envelope would require 1495m more than the node's
entire allocatable budget after replacing the existing 60m request. Do not
recommend that resize or treat reservation arithmetic as measured spare CPU.
The existing portfolio record now references the current observation without
refreshing stale utilization metrics or inventing financial attribution.
## Compare the constrained workload and propose a qualified allocation
```task
id: RESOURCE-WP-0008-T02
status: progress
priority: high
```
Consume the in-progress isolated app-container comparison at the existing
1000m CPU and 1Gi memory limits. The 60m request is a contention share/reservation,
not a CPU cap. Compare complete latency/error and CPU/throttle/memory evidence,
retain synthetic pacing assumptions and configurable safety factors, and state
what the sample cannot establish about production contention and shared services.
Reconcile any proposed request against a fresh node observation, release surges,
host operations, zero-request workloads and owner demand. An unchanged prototype
allocation may remain provisional; a larger allocation needs evidence that the
whole node budget fits. Do not infer human workflow acceptance.
## Return accountable changes and remaining measurement owners
```task
id: RESOURCE-WP-0008-T03
status: todo
priority: medium
```
Return the reviewed proposal/keep decision to CUST-WP-0071, naming exact receiving
workload/cluster records for every remaining change or measurement gap. Keep
unknown namespace ownership and missing demand explicit; do not mark this plan
finished while actionable leftovers exist only in prose. All application of
accepted sizing stays with existing runtime owners and their reviewed deployment
paths. This record does not create a parallel scheduler or change financial
authority.