diff --git a/REPORT.md b/REPORT.md index bc07c22..22744e4 100644 --- a/REPORT.md +++ b/REPORT.md @@ -1,132 +1,132 @@ -# REPORT — TASK-G: arm the HP install ISO + assisted-picker filing (2026-07-21) +# REPORT — TASK-H: HP node access + docs + today's findings (2026-07-21, evening) -Baseline `0073691`, clean tree; ISO tooling unchanged since v1.22.0, so this is a **rebuild at -v1.22.0** with a new profile name (`hp-demo`), not a version bump. +Baseline `59226ed`, clean tree. No product code touched; the N100 and guest 9201 were not modified. | Part | Status | |---|---| -| 1 — arm the ISO | **DONE**, filter verified from inside the built ISO | -| 2 — write it (gated) | **DONE**, byte-proven, and re-verified from the written stick | -| 3 — filing | **DONE** — **R-58** | +| 1 — tailscale on demo-hp | **DONE**, one item open (key expiry — needs an admin-console toggle) | +| 2 — inventory + docs | **DONE** — `documentation/operations/nodes.md` | +| 3 — findings filed | **DONE** — **R-59, R-60, R-61** + positive evidence recorded | -## Part 1 — the armed ISO +## Part 1 — tailscale on demo-hp -Built from an **uncommitted** working profile, `/mnt/5_hdd/felhom.eu/felhom-iso/hp-demo.profile` -(mode 0600, the `n100-demo.profile` precedent — a real target serial never goes in git). It is the -committed `profiles/hp.profile` with exactly one change: the placeholder serial replaced. +`tailscale 1.98.9` (Debian trixie apt repo, host package — same shape as `felhom-pve`). ``` -felhom-pve-9.2-1-v1.22.0-hp-demo-generic.iso -sha256 de693f622c4c628d75eade691b2e8cb008175231f3ed97e29535e61de12bbd0b -bytes 1 705 338 880 -loader shim (stock MS-signed chain; Secure Boot OK) secret-bearing: no +demo-hp 100.76.96.79 online direct path 37.191.56.193:45127 ``` -`validate-answer` gated on its **output text**, not `$?` — log shows -`The answer file was parsed successfully, no errors found!` through `gate_validate_answer`. +Verified **from DooPlex over the tailnet, no jumphost**: ping ~40 ms, `ssh root@100.76.96.79` works, +and the path is **direct**, not a DERP relay. `~/.ssh/config` gains `demo-hp` (tailnet) and +`demo-hp-lan` (`192.168.0.87` via `ProxyJump felhom-pve`) as the away-fallback. -### The filter, extracted from INSIDE the built ISO +The pre-auth key was passed as `--authkey=file:` and shredded immediately, so it never entered +the box's process list. -Not read from the profile — from the artifact, via `osirrox`, comments stripped: +### OPEN — key expiry is NOT disabled -``` -[disk-setup] -filesystem = "ext4" -filter-match = "all" -filter.ID_SERIAL_SHORT = "182195804614" -``` +`demo-hp` expires **2027-01-17**; `felhom-pve` and `dooplex` both have expiry disabled, so this is +the one place it diverges from the fleet convention. **I could not do it from here:** disabling +per-device key expiry is an admin-console toggle or an API call, and the supplied `tskey-auth-…` +pre-auth key cannot drive the API (verified — `GET /api/v2/tailnet/-/devices` returns **401**). -| assertion | result | -|---|---| -| pins the SanDisk X600 128GB (`sda`) `182195804614` | 1 occurrence | -| NVMe `58BS11AFT8MQ` present anywhere in the answer file | **0** | -| `REPLACE-WITH-TARGET-SSD-SERIAL` placeholder remaining | **0** | -| `filter.*` directives | exactly **1** | -| `disk-list =` directive | **0** | +**One click when convenient:** Tailscale admin → Machines → `demo-hp` → *Disable key expiry*. Or +hand me a `tskey-api-…` token and I will do it. Verify with: +`ssh demo-hp 'tailscale status --json' | python3 -c "import sys,json;print(json.load(sys.stdin)['Self'].get('KeyExpiry'))"` → should print `None`. -`filter-match = "all"` with a single `ID_SERIAL_SHORT` is an exact-one-disk selector. **The 1TB NVMe -is outside the filter** and is not touched by the installer; it joins later through the normal -Tárhely flow. +### A rule I broke, then fixed -*Method note:* my first sanity check grepped the whole profile and reported the NVMe serial as -present — it appears in a **comment** documenting what is excluded. Re-verified by sourcing the -profile and inspecting the consumed `FELHOM_DISK_SETUP` variable, then again from inside the ISO, -which is what actually ships. Grepping a file is not the same as reading what a tool consumes. +The join omitted `--accept-dns=false`, and MagicDNS immediately rewrote `/etc/resolv.conf` to +`nameserver 100.100.100.100` — exactly what `operations/tailscale.md` forbids for a host node. -## Part 2 — the write +Nothing broke *at the vacation site*: there is no pi-hole there, and both `gitea.dooplex.hu` and +`hub.felhom.eu` resolve publicly anyway. **The damage would have surfaced silently when the box comes +home**, where split-horizon is what makes `gitea.dooplex.hu` resolve to `192.168.0.180` locally — a +class of failure that looks like "the network is slow" rather than "DNS is wrong". +`tailscale set --accept-dns=false` restored `nameserver 192.168.0.1`; tailnet and agent unaffected. +The doc now carries the evidence and the instruction to pass the flag **at join time**. -Same procedure as TASK-F, no shortcuts. Full evidence block on the Samsung stick -(`/dev/disk/by-id/usb-Samsung_Flash_Drive_0375123010001586-0:0` → `/dev/sdc`, 119.5 GiB): +## Part 2 — the node inventory -| check | result | -|---|---| -| `ID_BUS=usb` | usb | -| `removable=1` | 1 | -| `findmnt` (device **and** every partition) | empty | -| `/etc/fstab` | 0 lines | -| LVM PV | 0 (only `/dev/sda3` is a PV) | -| ZFS | 0 pools on the host | -| guest configs referencing it | 0 | -| open handles | 0 | +**`documentation/operations/nodes.md`** (new). The fleet is now two hosts, both agent 0.92.1: +`demo-felhom-8363b5` (N100) and `demo-hp-bb76ea` (HP t740). Both are at the vacation site and travel +home ~2026-08-02. -Destroyed: the `hp-generic` **safety** ISO written in TASK-F. Excluded: `/dev/sdb`, the ADATA -HD710 PRO — mounted at `/mnt/hdd_1` and bind-mounted into guest 9201, i.e. **live customer data**; -it reports `ID_BUS=ata` and `removable=0`, so it fails both required positive checks despite a -`usb-…` by-id name. Operator confirmed the printed device; identity was **re-asserted at write -time** in case of re-enumeration since the confirm. +**demo-hp:** HP t740 Thin Client, s/n `8CN944035T`, AMI M42 v01.10 (11/11/2020), Ryzen Embedded +V1756B (8 threads), **30 GiB RAM**, PVE 9.2.2 as node `felhom-host`, guest 9201 `demo-hp` running, +WireGuard `10.77.0.3/32` up. -``` -device sha256: de693f622c4c628d75eade691b2e8cb008175231f3ed97e29535e61de12bbd0b -iso sha256: de693f622c4c628d75eade691b2e8cb008175231f3ed97e29535e61de12bbd0b -RESULT: MATCH - byte-proven -``` +| device | serial | role | +|---|---|---| +| `sda` SanDisk X600 128GB | `182195804614` | system disk — PVE + LVM + guest volumes | +| `nvme0n1` Toshiba KXG50 1024GB | `58BS11AFT8MQ` | **PRESENT AND UNENROLLED — do not touch** | -`dd … oflag=direct conv=fsync`, then the ISO's exact byte-length read back from the device and -hashed. Customer data drive re-checked after the write: still mounted at both paths. +The NVMe still holds its previous **NTFS** partition, is unmounted, and is in no LVM PV and no ZFS +pool — the exact-serial filter did its job on hardware it had never seen. It joins later through the +**Tárhely flow, never the installer**. -**Belt-and-braces:** the stick was then mounted read-only and the filter read back off the physical -device — `filter.ID_SERIAL_SHORT = "182195804614"`, NVMe serial absent. The profile is what I -intended, the ISO is what I built, and the stick is what you will boot; all three now agree. +**NIC map** (documented in both `nodes.md` and `scripts/iso/README.md`): `enp1s0f0–f3` are the +4-port `igb` card with **no carrier and no DHCP** at this site; **`enp2s0f0`** (`r8169`, MAC +`7c:d3:0a:77:d9:76`) is the onboard port that works and is now `vmbr0`'s bridge-port. MACs for all +five are in the doc. -Staged ISO copies removed from `felhom-pve:/root/` (71G free). +**Loader/firmware finding:** the box installed with the **shim** loader and **Secure Boot ENABLED** +(`mokutil --sb-state` → `SecureBoot enabled`). That retires an assumption — SB-off was an +N100-firmware workaround, not a Felhom requirement. -## Part 3 — R-58, assisted disk-picker install mode +**Access, and why it is awkward:** no operator SSH key is baked (the HP profile deliberately left +`FELHOM_ROOT_SSH_KEY` blank), so authentication is the **G1 break-glass root password vaulted in the +hub** (`host_recovery` row `demo-hp-bb76ea`, set 2026-07-21 16:24 UTC). I retrieved it by streaming +`/data/hub.db` out of the hub pod, extracting one field to a 0600 file, and **shredding the copy +immediately** — that DB holds every host's secret. Recipe is in the doc. This lockout is R-61. -Filed with your argument verbatim. Shape: a third ISO mode beside unattended-serial and -match-nothing-safety — enumerate candidates with size/model/serial, exclude the install media, take -a selection plus a confirm, proceed. Unattended+serial stays the appliance/factory mode. +**The operator-lab exception is documented prominently**, in both `nodes.md` and `tailscale.md`: +demo-hp is *customer-shaped* but tailscale is **not** part of that shape. Real customer boxes get the +WireGuard tunnel and the H1 OOB path and nothing else. A future product-shape audit finding tailscale +here must not conclude the product ships it. -**Slice 1 is the abort screen**, deliberately, because it is the same enumeration code and it -collapses the two-boot dance on its own: on filter-no-match, print the candidate table plus the -one-line hint naming which serial to put in the profile. Today the fail-safe is correct but mute — -which is exactly what made this task need a hand-read serial between two boots. +## Part 3 — findings filed -It matters most for the **BYO / reinstall** flow: a customer's existing hardware, or a rebuild of a -box whose disk layout nobody recorded, is precisely where the serial is unknown and a wrong guess is -destructive. +**TASK-G Part 3 verified as already filed** — R-58 (assisted disk-picker) exists and includes the +abort-screen candidate-table slice. Nothing to complete. -Also documented `scripts/iso/README.md` § **HP mini profile**: shim proven on this board by the -safety boot, the uncommitted-armed-profile pattern, verify-from-inside-the-ISO, and the prior-LVM -caveat below. +New rows: -## For the operator — boot instruction +- **R-59 [P1] — a no-DHCP install must HARD-ABORT.** Instead it baked `192.168.100.2` as a *static* + `vmbr0` address and completed: the install "succeeded", the box looked finished, and it could never + call home. The worst silent onboarding failure shape there is. The philosophy already exists one + layer over — the disk filter refuses loudly and touches nothing when it cannot identify its target; + networking should do the same, with the candidate NIC table on screen (same grammar as R-58 slice 1). +- **R-60 [P2] — first-boot NIC sweep self-heal.** Hub unreachable ⇒ DHCP across every carrier-bearing + NIC before settling. Today's repair was a human moving one cable; a sweep would have healed it + unaided. Scoped to first boot and the hub-unreachable condition only — a running box must never + re-shuffle its own networking. +- **R-61 [P1] — the baked root password must be knowable.** The ISO mints a throwaway hash per build + and discards the plaintext, so nobody can reach the console of a box they just installed. Slice 1: + emit it into the build REPORT + operator cheat-sheet alongside the sha256. **A fixed well-known + password is explicitly rejected** (operator ruling) — a pre-pairing box sits on a stranger's LAN. -**Stick → HP, boot it, walk away.** It formats the SanDisk `sda`, installs, and reboots into day-0. -Secure Boot stays **ON** (this is the shim loader, proven on this board). Nothing else to press. +Positive evidence, same session: -**One thing that could stop it, worth knowing before you walk away.** The Proxmox auto-installer does -not force-clear a pre-existing **LVM PV** signature. If that SanDisk carries a previous -Linux/Proxmox install, the install aborts at *"cannot init physical volume"* — safely, touching -nothing. A previous Windows/NTFS install is fine. If you see that message: wipe the target -(`blkdiscard -f /dev/sda` from any live environment) and boot **the same stick** again — the ISO does -not need rebuilding. This is documented behaviour (spike S2b), not a defect in this build. +- **R-21 slice C is no longer a one-board result.** The row and the capability-map ISO row now read + **PROVEN-LIVE on TWO different boards** (N100 2026-07-18; HP t740 2026-07-21) — virgin hardware, + one pass, self-registration → operator bind → day-0 → running guest + agent check-in. The Secure + Boot finding is recorded there too. +- **Fresh-box floor lift**: the new box came up on golden **0.153.0** and self-updated to the fleet + floor **0.156.0** during day-0, unattended. Cited on the publish-train capability-map row — the + floor mechanism works on first contact, not only on boxes with history. +- **The t740 five-NIC trap** is recorded as a board gotcha in `scripts/iso/README.md`. -## Remaining (unchanged from TASK-F unless noted) +## Observations -- **Agent v0.93.0 is source-only** — committed, unbuilt, unpublished; rides the next agent train. -- **R-17** — deleting the `u629193-sub1` archive is an operator **console** action. -- **The full catalog resolvability table** is still owed (needs one `docker login` on DooPlex). -- **wanderer** was re-pinned to a 3-service upstream shape and has never been deployed. -- **`hidden` lifecycle state** has no catalog user yet; only `abandoned` is exercised end to end. -- **After the HP comes up:** bind it on the hub and run day-0; the 1TB NVMe is added afterwards - through the Tárhely flow, never through the installer. +1. **The break-glass path works, and it is also the argument for R-61.** Reaching a box whose root + password was never known required a working hub, a working network, and operator tooling — at + exactly the moment the reason you want the console is usually that one of those is broken. +2. **`apt update` fails on this box** against the PVE **enterprise** repos (401, no subscription) — + pre-existing from the install, unrelated to today. I worked around it with a list-scoped + `apt-get update` rather than editing the box's repo config. Worth deciding whether `host-install` + should switch fresh boxes to the no-subscription repo; left alone deliberately. +3. **A heredoc silently ate a piped secret.** `printf … | ssh host 'bash -s' <<'EOF'` sends the + *heredoc* as stdin, so the piped key never arrives and `read` consumes script text instead. The + working shape is the command as an argument, with stdin free for the secret — then stage it as a + file and use `--authkey=file:`. Worth remembering next time a credential has to cross an SSH hop.