Repository navigation
docs(rack): put the KVM on a shelf, decide the KVM and PDU maps, retire two bought rows - #1198
Merged
Merged
Conversation
…re two bought rows The MT-VIKI 801UK-L is a half-U device: its ears take two of a unit's three holes and sit about 8 mm off centre either way, so it goes on a third 1U vented shelf at U23. hardware.md gains a Console and power maps section: KVM ports in the runbook's power-up order, PDU outlets bottom of the rack up with the bricks together at the far end. The keyboard drawer and ifrit's Tiny mount, mounted 2026-10-15, leave Everything still to buy; twelve rows become ten and README's count follows. The move runbook cables the KVM and the PDU by the maps instead of asking for them at the rack. Refs #919. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
There was a problem hiding this comment.
🟡 Changes recommended
The power instructions require a nonexistent UPS main group and leave both laptop chargers unassigned.
2 open findings
What changed in this PR
This documentation-only PR prepares the 32U rack move by defining cabling maps and updating hardware and purchase records.
Changes:
- Adds a third vented shelf for the KVM.
- Defines KVM port and PDU outlet assignments.
- Removes two purchased items from the roadmap and updates README’s count.
| File | Description |
|---|---|
| README.md | Updates outstanding purchases from twelve to ten. |
| docs/runbooks/move-into-the-32u-rack.md | Updates mounting, shelf requirements and cabling instructions. |
| docs/roadmap.md | Removes bought mounts and adds the KVM shelf. |
| docs/hardware.md | Records mounted parts and console/power maps. |
| docs/changelog.md | Records the rack-planning decisions. |
🧠 Review effort: Balanced
💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.
Gerrrt
enabled auto-merge (squash)
October 15, 2026 23:26
…groups, not assume a main one Review on #1198: the PDU map omitted both laptops' chargers, which §1.7 counts, and asked for the PDU to be on a UPS "main group" that may not exist. The chargers go on the UPS's own outlets beside the TP-Link in either case, so the map holds whether the DL360 has one supply or two, and §1.7's count becomes 8–9 of 10. The check now reads which outlet group feeds the PDU, the TP-Link and the chargers, and compares each group's switching settings with ADR-0049 and §7's power-up order. Refs #919. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.


What changed
Docs only, no change to the running lab. The KVM at U23 goes on a third 1U vented shelf, because the MT-VIKI 801UK-L is a half-U device whose ears take two of a unit's three holes and sit about 8 mm off centre either way.
hardware.mdgains a Console and power maps section: KVM ports in the runbook's power-up order (morpheus1,trinity2,Saruman3,smaug4 if its rear VGA is used,ifrit5), and PDU outlets bottom of the rack up with the bricks together at the far end. The keyboard drawer andifrit's Tiny mount, both mounted on 2026-10-15, leave Everything still to buy; twelve rows become ten, and README's count follows. The move runbook's §0.1, §6 and §6.1 say the same, and §6.1 now cables the KVM and the PDU by the maps instead of asking for them to be written down at the rack. Changelog entry dated 2026-10-15.Why
Refs #919. The runbook asked for the KVM port map to be written down during the window; deciding it before cabling is cheaper, and the PDU map had never existed. The two stale purchase rows broke the roadmap's own rule that a bought part leaves the table.
Blast radius
None: documentation only.
secrets/*.sops.yamlVerification
python3 scripts/check_docs.py:docs OK — 158 Prometheus + 20 Loki rules, 8 dashboards, 156 panels, 12 assertions, after the README count was corrected from eleven to ten../scripts/lint.sh: every check PASS.make validate:all checks passed (5 skipped — NOT run, not verified). The five skips are the worktree ones: the offsite path refusal (CI covers it), unit syntax, installed timers, the sensitive stack's timers and the homelab-* units, none of which this checkout serves. Lint ran without--require-all; CI runs the hardened boots, the secret scan and the close-keyword check too.python3 scripts/check_close_keywords.py --texton this body: clean.make validatepassesDeployed to the lab and confirmed working (nothing to deploy)
Docs updated (
docs/, service README, ordocs/roadmap.md)🤖 Generated with Claude Code