Repository navigation
Description
Storage assembly & design validation
Purpose
The sketched storage tiers become real, tested pools with per-drive records.
Scope
Validate then build: HDD tier 3 hosts x 4x16 TB RAID 10 for RustFS cold buckets; fast NVMe tier (all Samsung 990s); mid tier striped x3 / mirrored x2 'RAID 100'. Install the 8 still-boxed HDDs and capture their baseline SMART/firmware data (NVMe baseline already exists on a backup drive). Per-drive records/passports so TALM and LINSTOR decisions have real data.
Execution mode
Physical: drive installation and machine assembly are hands-on; design validation, benchmarks, and record-keeping online. Physical tasks must state location, tools, and verification (photo/SMART output).
Dependencies
Depends on epics 1 (selection) and 3 (power). Feeds epics 6 and 8.
Acceptance
Design validated or revised in writing; first-node drives installed with baseline data recorded; pool layout per first node committed as config, not folklore.
Structure
Part of the basement cluster program. Project home: planning/roadmaps/basement-cluster.md in this repo. Full program register: goal workspace files/idea-register-2026-10-20.md (Muse goal goal_ba17a5777c21). Detailed task issues are filed as sub-issues of this epic; dependencies use blocks / blocked-by links.
Hard rules for the whole program: no HashiCorp anything; everything bootstrappable from config, no hand-built state.
Activity
developing-today commentedon Oct 27, 2026
CODE-226
dezren39 commentedon Oct 27, 2026
Epic map — basement cluster program
This epic: #1085 — Storage assembly & validation
Dependencies: Blocked by #1080, #1083; feeds #1086, #1088
Filed in the 2026-10-26 cluster seed. Project home:
planning/roadmaps/basement-cluster.md.dezren39 commentedon Oct 27, 2026
Predecessor: this epic supersedes dezren39/darwin-nix#504 (filed 2026-10-20 in the nix repo during the program's first filing pass). A superseded-pointer comment is on that issue; work happens here.
dezren39 commentedon Nov 3, 2026
agent-note:v1
state: triaged (coverage pass)
Coverage pass 2026-11-03: reviewed in the fleet-wide per-issue pass (drift / overlap / already-handled / stale / blockers / ordering). Verdict: still live; queued in repo-health-code-pilot as triage at priority 30 (priority/none). If this is already handled at current main, the triage closes it with evidence; if it needs splitting, the triage files the split.
Found via: coverage pass 2026-11-03
dezren39 commentedon Nov 3, 2026
Queue note (triaged, task 7642): verdict — still live, and not queue-workable yet. It is gated by the Now bucket and by hands-on hardware work.
What it is
Roadmap
Nextbucket (planning/roadmaps/basement-cluster.md): build the HDD RAID 10 pools and NVMe tiers, validated against per-drive records — 3 hosts × 4 × 16 TB in RAID 10 for RustFS cold buckets, an all-990-Pro NVMe fast tier, a striped-and-mirrored mid tier, plus the DRBD + LINSTOR replication design from epic 5, including the quorum/witness question (spare PC or Pi as tie-breaker).Why it cannot be worked from the queue right now
status/needs-triagequeue/claimedby another laneuser/owned,mode/physicalTwo further gates are inherent, not queue policy: the epic is
mode/physical— it needs hands on the drives (cable, label, build pools) — and its acceptance requires per-drive records and a node-loss replication test, neither of which a code worker can perform.Gap found
subIssuesfor #1085 is empty, as for #1088: the roadmap names the epic but no child tasks are filed. Filing children now would be premature — the pool layout per node depends on #1080's drive inventory. The right first child is the per-drive record format plus the inventory table, and it should be filed once #1080 closes.Verdict
Keep open as roadmap content; do not re-queue for triage. Labels set to
status/blocked+queue/blocked(wasstatus/needs-triage+queue/queued).Unblock condition: #1080 selects the first nodes and #1083's power/space work is done, then a human-supervised session builds and validates the pools on the physical machines.