TheRON/mechanical-compiler
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Current state

Browse Source 
Live state of the Mechanical Compiler staging instance on `srv-b`.
This commit is contained in:
TheRON
2026-08-16 12:01:20 -04:00
parent a00ba35451
commit eac601ed99
1 changed files with 299 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/STAGING-STATE.md
+299
View File
@@ -0,0 +1,299 @@
# STAGING-STATE.md
Live state of the Mechanical Compiler staging instance on `srv-b`.
| | |
|---|---|
| Updated | 2026-08-15, after network isolation |
| Instance | Staging / development |
| Specification | `ENVIRONMENT.md` revision 5 |
| Failure log | `FAILURES.md` |
| Method | Manual, one command group at a time |
---
## 0. How to use this file
This is the authoritative record of what is true on `srv-b`. Where it and
`ENVIRONMENT.md` disagree, **this file wins for facts** and the specification
is defective and must be corrected.
Completed and remaining work are in the same document deliberately. They are
two halves of one boundary; separating them guarantees they drift.
Before any command: read this file, confirm the immediately relevant live state
with a read-only command, then issue one command group. If it fails, record it
in `FAILURES.md` before changing anything else.
Production will get its own `PRODUCTION-STATE.md`. The specification is shared;
the state is not.
---
## 1. Instance values
These are the `srv-b` bindings for the parameters in `ENVIRONMENT.md`.
### Host
```
hostname srv-b / srv-b.dev.infra
platform Proxmox VE 8.4.0, Debian 12, kernel 6.8.12-9-pve
hardware HP ProLiant DL360 G7, 2 x Xeon X5650, 24 threads, 31 GiB
storage P410i, 4 x EG0146FAWHU, RAID 1+0, all SMART OK
local directory /var/lib/vz, ~70 GiB free iso,vztmpl,backup
local-lvm LVM-thin pve/data, 166.9 GiB rootdir,images
vmbr0 10.0.0.12/24 on enp3s0f0, gw 10.0.0.1 management, LAN
vmbr1 10.20.0.1/24, bridge-ports none service, portless
wg0 10.110.0.12/32, peer wg-pk.civicus.us:51820, allowed 10.110.0.0/22
resolver 75.75.75.75, search dev.infra
spare NICs enp3s0f1, enp4s0f0, enp4s0f1 — unconfigured, deliberately
template local:vztmpl/debian-12-standard_12.12-1_amd64.tar.zst
```
### Containers
| | CT 100 | CT 101 |
|---|---|---|
| hostname | `mechcomp` | `mcproxy` |
| role | application, worker | reverse proxy, TLS |
| features | `nesting=1,keyctl=1` | `nesting=1` |
| cores / RAM / swap | 16 / 16384 / 4096 | 2 / 1024 / 512 |
| rootfs | 40 GiB `local-lvm` | 8 GiB `local-lvm` |
| `mp0` | 60 GiB → `/var/lib/mechcomp`, `backup=0` | — |
| `net0` | `eth0` on `vmbr1`, `10.20.0.10/24`, gw `10.20.0.1` | `eth0` on `vmbr1`, `10.20.0.11/24`, gw `10.20.0.1` |
| Debian | 12.15 | 12.15 |
**Neither container has a LAN interface.** See F-017.
### Names and identity
```
mechanical-compiler.dev.infra -> 10.20.0.11 (CT 101, the proxy)
mechcomp.dev.infra -> 10.20.0.10 PVE-generated
mcproxy.dev.infra -> 10.20.0.11 PVE-generated
ADMIN_USER sandor
uid 1000, groups sandor + mechcomp(996)
shell /bin/bash, both containers
authorized key SHA256:2pNffCscUUW5Wbs9uMepLEvMLKPqUV/7Lpk5tw7iWSY
root@srv-b — bastion pattern, see §4
service user mechcomp, uid 999, gid 996
home /var/www/mechcomp, shell /usr/sbin/nologin
```
### Host NAT, as persisted in `/etc/iptables/rules.v4`
```
nat POSTROUTING
-s 10.20.0.0/24 -d 10.0.0.0/24 -j RETURN # LAN not translated
-s 10.0.0.0/24 -o wg0 -j MASQUERADE # pre-existing
-s 10.20.0.0/24 -o wg0 -j MASQUERADE # containers -> WireGuard
-s 10.20.0.0/24 -o vmbr0 -j MASQUERADE # containers -> internet
filter FORWARD
-s 10.20.0.0/24 -d 10.20.0.0/24 -j ACCEPT # container <-> container
-s 10.20.0.0/24 -d 10.0.0.0/24 -j DROP # LAN blocked
```
Rule order is load-bearing. `RETURN` must precede both masquerades.
### TLS
```
CA CN = Mechanical Compiler Staging CA (locally generated)
leaf CN = mechanical-compiler.dev.infra
SAN = DNS:mechanical-compiler.dev.infra
validity 2026-08-16 -> 2028-11-18 <-- expires, nothing renews it
trusted srv-b, CT 100, CT 101
key mode 0600 on CT 101
```
No Kane County Civic Infrastructure CA issuance path exists on `srv-b`: no
trust anchor, no `step`, no `cfssl`, no EasyRSA. Proxmox's own CA was
deliberately not reused. Replacing this leaf later is two file copies and a
reload.
### Application environment
`/etc/mechcomp/mechcomp.env`, `root:mechcomp`, `0640`:
```
MECHCOMP_ENV=staging
MECHCOMP_BIND=10.20.0.10
MECHCOMP_PORT=8770
MECHCOMP_BASE_URL=https://mechanical-compiler.dev.infra
MECHCOMP_DATA_DIR=/var/lib/mechcomp
MECHCOMP_LOG_LEVEL=info
MECHCOMP_DB_URL=sqlite:////var/lib/mechcomp/db/mechcomp.sqlite3
MECHCOMP_WORKER_CONCURRENCY=4
MECHCOMP_CAD_BACKEND=none
MECHCOMP_ARTIFACT_RETENTION_DAYS=30
MECHCOMP_SECRET_KEY=<generated, not recorded>
```
### Rollback copies on disk
```
srv-b /etc/network/interfaces.before-mechcomp
/etc/hosts.before-mechcomp
/etc/hosts.before-svcfqdn
/etc/iptables/rules.v4.before-svcnat
/etc/iptables/rules.v4.before-lanblock
/root/pct-100.before-svcnet
/root/pct-101.before-svcnet
CT 100 /etc/hosts.before-svcfqdn
CT 101 /etc/hosts.before-svcfqdn
```
---
## 2. Verified
Each line was demonstrated by command output, not inferred.
### Host and network
- [x] `vmbr1` created, active, `10.20.0.1/24`, portless
- [x] Duplicate address detection run before each container creation
- [x] Container → internet reachable (`deb.debian.org`, `gitea.barternetwork.us`, both 200)
- [x] Container → WireGuard network reachable (`10.110.0.12`)
- [x] **Container → LAN unreachable** — the isolation requirement
- [x] Container → `srv-b` reachable at `10.0.0.12` (INPUT path, required)
- [x] Container ↔ container reachable over `vmbr1`
- [x] LAN → Proxmox console unaffected (`10.0.0.12:8006` → 200)
- [x] Network configuration persisted, survives reboot
### CT 100
- [x] Debian 12.15, systemd running, zero failed units, no pending upgrades
- [x] `en_US.UTF-8` generated and active
- [x] `/var/lib/mechcomp` is a real separate ext4 filesystem
- [x] Service user `mechcomp`, home `/var/www/mechcomp`, writable
- [x] Application directory layout created with correct ownership
- [x] Base packages installed; OpenSCAD / Qt / X11 absent — verified `clean`
- [x] Docker working, `overlay2` / `systemd`, no `fuse-overlayfs` needed
- [x] Repository cloned at `e85c4f4e`, verified as the owning user
- [x] Python 3.11.2 venv created, owned by `mechcomp`
- [x] `mechcomp.env` written with generated secret
- [x] `openssh-server` installed, enabled, active
- [x] Application listener bound to service network only — LAN-side bind refused
### CT 101
- [x] Debian 12.15, systemd running, zero failed units after `nesting=1`
- [x] `en_US.UTF-8` generated and active
- [x] nginx 1.22.1 installed, `nginx -t` passes
- [x] Debian `default` site removed; only the project vhost is enabled
- [x] Local CA created, leaf issued, trusted on all three hosts
- [x] HTTP → HTTPS redirect, TLS termination, proxy to `10.20.0.10:8770`
- [x] `openssh-server` installed, enabled, active
- [x] **`X-Forwarded-Proto: https` observed at the backend** — needs re-proof, see §3
---
## 3. Remaining — infrastructure
In dependency order. Application-dependent work is in §5.
### Immediate
- [ ] **`mechcomp-placeholder.service`** — recreate the backend as a supervised
unit. It died on the F-017 reboot and nothing currently listens on 8770.
Everything below that touches the proxy depends on this. See F-019.
- [ ] **Re-prove `X-Forwarded-Proto`** end to end. The earlier proof was taken
when clients were on `10.0.0.x`; header values will now read `10.20.0.x`.
Re-establish rather than assume it survived the topology change.
- [ ] **`default_server`** on the CT 101 vhost, both `listen 443` lines. Two
words. Prevents catch-all behaviour depending on file ordering once a
second server block exists. See F-012.
### Mail — blocks alerting
- [ ] **Postfix relay.** Currently `inet_interfaces = loopback-only`,
`relayhost` empty, no `root:` alias. Alerts go to a mailbox nobody opens.
- [ ] **`root:` alias** to a real destination.
- [ ] Needs: address of the `wg-pk` relay, and whether it accepts
unauthenticated from `10.110.0.12`. **Open question for CIVICVS.**
### Monitoring and backup
- [ ] **`smartd`** on `/dev/sda -d cciss,0` through `cciss,3`. Depends on mail.
- [ ] **`vzdump` job** — both containers, `local`, snapshot, zstd, 02:30.
`mp0` already carries `backup=0`.
- [ ] **First `vzdump` run — report the archive size.** Retention is a
placeholder `keep-last=3` until this number exists.
- [ ] **Free-space guard** — refuse and alert below 20 GiB on `/var/lib/vz`.
- [ ] `/var/lib/vz/mechcomp-app/` and the host-side pull script.
- [ ] `mechcomp-backup --stdout` in CT 100. Structure can be built and tested
against the current data directories without application output.
### Gold media
- [ ] 32 GB USB stick: LUKS2, ext4, label `MC-GOLD-01`, `/mnt/gold` `noauto`.
- [ ] `gold-archive.sh` with `SHA256SUMS` verification before unmount.
- [ ] **Open:** where the 3+ TB USB disk is attached now that `annales` is out
of scope. Until known, stick-to-disk copying is a manual step.
---
## 4. Access model
Confirmed by CIVICVS: no workstation access from the home LAN is required.
```
internet -> WireGuard -> srv-b -> containers
```
`srv-b` is the bastion. The authorized key is `root@srv-b`, which is
consistent: root on the host can `pct enter` regardless, so SSH to the
containers adds no privilege. It does mean **every path to a container runs
through `srv-b`** — deliberate, not a limitation.
If direct WireGuard-side access to the catalogue is wanted later, that is a
route addition for `10.20.0.0/24` on the hub. Not now.
No DHCP anywhere. Two containers with fixed addresses on a portless bridge is
the entire address space; a DHCP server would add a daemon, a lease database
and a failure mode in exchange for nothing.
---
## 5. Blocked on application code
None of this can be honestly completed while the repository is `LICENSE` and
`README.md`. It is not provisioning work and should not be attempted as such.
- [ ] `requirements-base.txt` / `requirements-cad.txt` and dependency install
- [ ] `mechcomp.service` and `mechcomp-worker.service`
- [ ] Reference toolchain image `mechcomp/reference-toolchain:8.0.0`
- [ ] Fixture reproduction against `ddd0f154…`
- [ ] Application-runtime acceptance
- [ ] Replacing the placeholder with the real service
The Shapely port is the architect's work item and gates all of the above.
---
## 6. Open questions
| # | Question | Blocks |
|---|---|---|
| 1 | `wg-pk` relay address; unauthenticated from `10.110.0.12`? | mail, `smartd`, backup alerts |
| 2 | Where is the 3+ TB USB disk attached? | gold redundancy step |
| 3 | First `vzdump` archive size | final retention value |
Question 3 is answered by doing. Questions 1 and 2 need CIVICVS.
---
## 7. Closed questions
| Question | Answer |
|---|---|
| Kane County CA issuance path | None on `srv-b`. Local staging CA generated instead. |
| `openssh-server` present | Yes, both containers. |
| `ADMIN_USER` / key | `sandor`; `root@srv-b` key, bastion pattern. |
| DHCP pool on `10.0.0.0/24` | **Not applicable.** Containers are no longer on that network. |
| Docker storage driver | `overlay2` / `systemd`. No fallback needed. |
| LAN workstation access | Not required. WireGuard through `srv-b`. |