docs(hardware): put the patch panel back above the PDU - #163
Merged
Conversation
The Rack table had U7 as the patch panel and U8 as the PDU. It is the other way round. Confirmed against the rack on 2026-08-29, after this table and the wiki's rack page were found to disagree about exactly these two slots and neither could settle it on paper. Everything else in the table is right, which is what makes this two swapped rows rather than a numbering problem: U1-U2 holds the UPS at the bottom and U9 the switch at the top, so higher U means higher in the rack, and every other row matches what is actually bolted in. Worth stating plainly that nothing was ever at risk. The rack order is load-bearing outside this repository -- the household's printed emergency card says to unplug "the box at the very top" and power-cycle "the 5th box from the top", meaning the switch and morpheus. Neither moves. The swap is between the second and third units, below the top slot and above the fifth, so the count to morpheus is unchanged under either version. Wrong documentation, not a wrong instruction. The correction is recorded under the table rather than applied silently, following the oracle entry in Compute. The two documents disagreed and this one lost; leaving no trace of that invites someone reconciling them again later to swap it back. Nothing in this repository read the order. check_docs.py cross-checks the Compute table against docs/network.md and does not look at the Rack table, and no diagram, runbook or config references the U position of either device. Verified: scripts/check_docs.py passes -- 35 Prometheus + 13 Loki rules, 5 dashboards, 84 panels, 6 assertions -- and markdownlint-cli2 reports 0 issues across 34 files. Closes #162 Co-Authored-By: Claude Opus 5 <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
Two rows in the Rack table in
docs/hardware.md.U7is the PDU andU8is the patch panel; they were listed the other way round. A short note under the table records the correction, following theoracleentry in Compute.Nothing about the running lab changes — this is documentation matching the rack.
Why
Closes #162. Confirmed against the rack on 2026-08-29, after this table and the wiki's rack page were found to disagree about exactly these two slots and neither could settle it on paper.
Everything else in the table is right, which is what makes this two swapped rows rather than a numbering-direction problem:
U1–U2holds the UPS at the bottom andU9the switch at the top, so higher U means higher in the rack, and every other row matches what is bolted in.The note is deliberate rather than a silent swap. The two documents disagreed and this one lost; leaving no trace of that invites whoever reconciles them next to swap it back.
Blast radius
None. Nothing in this repository reads the rack order —
check_docs.pycross-checks the Compute table againstdocs/network.mdand never looks at the Rack table, and no diagram, runbook or config references the U position of either device.Worth stating plainly that nothing was ever at risk outside the repo either. The rack order is load-bearing for the household's printed emergency card, which says to unplug "the box at the very top" and power-cycle "the 5th box from the top" — the switch and
morpheus. Neither moves. The swap is between the second and third units, below the top slot and above the fifth, so the count tomorpheusis unchanged under either version. Wrong documentation, not a wrong instruction.secrets/*.sops.yamlVerification
Ran both CI checks locally against the change:
make validatepasses —check_docs.pyandmarkdownlint-cli2as above; this change touches no compose or secrets inputdocs/, service README, ordocs/roadmap.md)Built in a scratch worktree off
origin/main, because the checkout on this machine is mid-change onclaude/ups-battery-proven-4299dcwith uncommitted work instacks/observability/compose.yaml. That branch and its working tree were left untouched.