Files
shopdb-flask/docs/INSTALL-WINDOWS-IIS.md
cproudlock aea2905de0
Some checks failed
CI / backend (push) Failing after 7s
CI / naming (push) Successful in 2s
CI / frontend (push) Successful in 9s
CI / migrations-mysql (push) Failing after 7s
fix(installer): stop it lying, stop it leaking, and make it findable
Nine fixes from a review of the installer against its actual audience: DT leads
at sister sites who are not Windows, IIS or Python specialists and who will lean
on an AI assistant to get through it.

TRUTHFULNESS. The preflight was advisory - an operator read 'IIS is not
installed', pressed Next, answered five more pages and the install died partway
through with Python already on the box. The results page now blocks while
anything is failing, repaints on every run instead of latching after the first,
and offers 'Check again' so a fixed problem does not mean starting over. On
failure the wizard said 'Nothing was left running', which is false in every path
because the stages run with -OnFailure never: it now says the server is
part-configured, that re-running is safe, and how to remove it. The final page no
longer reads 'ShopDB-Flask is ready' after a failed install.

SECRETS. The generated MySQL root password went to Write-Host in a process the
wizard runs hidden - so nobody saw it - and stdout is forwarded into the setup
log operators are told to send to support, so it was permanently recorded for
everyone who did not need it. It now goes to an ACL'd file. Database dumps, which
contain every user password hash, landed in a ProgramData directory readable by
every user on the box; the directory is now locked at creation.

UPGRADES ON REMOTE-DATABASE SITES. mysqldump was looked for only under local
MySQL install paths, so a site whose database is on another host silently skipped
every pre-upgrade backup - after stage 2 had already stopped the pool and
replaced the tree. Find-MysqlTool now prefers a client shipped in the bundle,
stage 2 stages it onto the server, preflight reports when it is missing, and
mysqlclient\ is an optional locked payload.

UNINSTALL. A subpath install is an IIS Application, not a site; removing only the
site left the application pointing at a deleted directory, so the parent site -
at West Jefferson, the live classic ASP - served 503 on that path forever while
Add/Remove Programs reported success. Uninstall now reads MOUNT_PATH and removes
the application. The firewall rule was created as "$SiteName $SitePort" and
removed as the literal 'ShopDB-Flask 8090', which matches nothing.

DAY-2 TOOLING. Every shortcut now passes -AppRoot and -SitePort, and the console
forwards them through its own elevation and 32-bit relaunches instead of
discarding them - a non-default directory or port made it report a healthy site
as broken, from a shortcut the installer wrote. 'Open ShopDB-Flask' resolved to a
hardcoded localhost:8090 that was wrong for every subpath install; it now asks
the console, which reads the address the installer recorded, and no longer
demands administrator to open a browser.

SMOKE TEST. The parent-site port lookup filtered for an http binding and
defaulted to 80, so an https-only parent site failed a working install with a red
dialog.

DOCS AND /api/docs. The installer was invisible: nothing in docs/, README.md or
CLAUDE.md mentioned it, so a DT lead or their assistant landed on the manual IIS
runbook and hand-built the very server the installer then refuses to upgrade.
docs/INSTALL-WINDOWS.md and docs/OPERATE-WINDOWS.md are now the canonical route,
the two manual runbooks are bannered as reference-only, README and CLAUDE.md
route by target, and llms.txt tells an assistant which document to follow and to
ask for 'check -Json' before diagnosing. Both ship on the server, along with
openapi.json and llms.txt - without those the self-hosted /api/docs was broken on
every installed box, which matters most to the sites least able to debug it.
Stage 5 now checks it actually serves.

shopdb-admin.ps1 gains 'check -Json': one structured, secret-free block covering
version, publishing method, IIS state, HTTP reachability, database, Python
version, plugins and errors. That is the cheapest useful answer to 'the operator
will ask an LLM' - it works with no infrastructure, which a install-time MCP
server could not.
2026-08-03 14:39:38 -04:00

255 lines
11 KiB
Markdown

# ShopDB - Windows + IIS install runbook
> **Not the route for a new site.** Sister sites install from the Windows
> installer — one `.exe`, no manual IIS work: **[INSTALL-WINDOWS.md](INSTALL-WINDOWS.md)**.
>
> This document is the **manual** procedure, kept for reference and for
> hand-built servers that predate the installer. Note that the installer will not
> adopt a server built this way without `-AdoptExisting`, on purpose.
A step-by-step, **tested** install for a new site on Windows Server / Windows 11
with IIS in front of the Flask app (HttpPlatformHandler -> waitress), backed by
MySQL. This runbook was validated end to end on a win11 + IIS + MySQL 5.6 box.
`APP_ROOT` below = the deploy folder, e.g. `C:\shopdb-flask` (where `wsgi.py`
lives). Run PowerShell as Administrator.
---
## 0. Prerequisites
| Need | Notes |
| --- | --- |
| **Python 3.14** (64-bit) | `python --version` |
| **IIS** with **HttpPlatformHandler** | https://www.iis.net/downloads/microsoft/httpplatformhandler (direct MSI: `download.microsoft.com/download/8/1/3/813AC4E6-9203-4F7A-8DD5-F3D54D10C5CD/httpPlatformHandler_amd64.msi`) |
| **MySQL 8.0** (standard for new installs) | reachable from the app host. 5.7+ is supported on an existing server; 5.6 needs the flags in step 1. CI and the container image both target 8.0. |
| URL Rewrite (optional) | only for the real-client-IP rule; skip it and the app still runs |
The app itself pulls in `waitress` and `tzdata` from `requirements.txt` (step 4).
---
## 1. MySQL: flags (5.6 only) + database + user
On **MySQL 5.6 only**, add to `my.ini`/`my.cnf` under `[mysqld]` and restart MySQL
(5.7+/8.0 need none of this):
```
innodb_file_per_table = 1
innodb_file_format = Barracuda
innodb_large_prefix = 1
```
Without them, `flask db upgrade` fails with **error 1071** ("key too long") - the
migrations use `ROW_FORMAT=DYNAMIC`, which needs the 3072-byte prefix these unlock.
Then create the database (utf8mb4) and an app user:
```sql
CREATE DATABASE shopdb_flask CHARACTER SET utf8mb4 COLLATE utf8mb4_unicode_ci;
CREATE USER 'shopdb'@'%' IDENTIFIED BY 'CHANGE_ME';
GRANT ALL PRIVILEGES ON shopdb_flask.* TO 'shopdb'@'%';
FLUSH PRIVILEGES;
```
---
## 2. Deploy the app files
Copy the release (the repo minus `venv/`, `.git/`, `node_modules/`,
`frontend/src/`) to `APP_ROOT`. It must contain `wsgi.py`, `shopdb/`, `plugins/`,
`migrations/`, `requirements.txt`, and the pre-built `frontend/dist/`.
---
## 3. Virtual env + dependencies
```powershell
cd APP_ROOT
python -m venv venv
venv\Scripts\python -m pip install -r requirements.txt
```
This installs Flask, SQLAlchemy, PyMySQL, **waitress** (the WSGI server IIS
launches) and **tzdata** (Windows has no IANA tz database; without it the
notifications plugin fails with "No time zone found with key America/New_York").
---
## 4. Secrets + connection (.env)
Create `APP_ROOT\.env` (read by `wsgi.py` via `load_dotenv()`). Lock its ACLs to
the app-pool identity + admins.
```
FLASK_ENV=production
SECRET_KEY=<64+ random chars>
JWT_SECRET_KEY=<another 64+ random chars>
DATABASE_URL=mysql+pymysql://shopdb:CHANGE_ME@<mysql-host>:3306/shopdb_flask?charset=utf8mb4
CORS_ORIGINS=http://<the site's own hostname-or-ip:port>
```
Generate a key: `venv\Scripts\python -c "import secrets;print(secrets.token_urlsafe(64))"`.
Production **refuses to boot** if any of `SECRET_KEY`, `JWT_SECRET_KEY`,
`DATABASE_URL`, `CORS_ORIGINS` is missing or a dev default.
---
## 5. Preflight (catch problems before installing)
```powershell
$env:FLASK_APP="shopdb"
venv\Scripts\flask db-utils preflight
```
Checks Python, required env, DB connectivity, and the MySQL 5.6 index flags, and
prints exactly what to fix. Fix any **FAIL** before continuing.
---
## 6. Schema + data + plugins + admin
```powershell
$env:FLASK_APP="shopdb"
venv\Scripts\flask db upgrade # creates every table (to head)
venv\Scripts\flask seed reference-data # statuses, machine/location/rel types
venv\Scripts\flask seed permissions
venv\Scripts\flask seed settings
# enable the plugins this site tracks (registry is empty on a fresh box).
# usb + employees install DISABLED by default - enable them later in the wizard
# if the site wants those (they create extra tables).
foreach ($p in "computers","machines","network","notifications","printers","knowledgebase","slides","warranty") {
venv\Scripts\flask plugin install $p
}
# first admin (password generated + printed once - store it):
venv\Scripts\flask seed admin --username admin --email admin@yourfacility.example.com
```
> Prefer no CLI? Skip `seed admin` (and even the seed steps): start the site, and
> the login page offers to **create the first admin** on a fresh instance, then
> the setup wizard can seed reference data. Either path works.
---
## 7. IIS site
Two supported deployment methods:
- **Method A - own site (recommended, default):** the app gets its own IIS
site, port (or hostname), app pool, and venv. Steps 1-5 below.
- **Method B - subpath under an existing site:** the app runs as an IIS
**Application** (e.g. `/ops`) under a site you already have (such as the
classic ASP site or Default Web Site), so it shares that site's binding and
TLS cert: `https://<host>/ops/`. Do steps 1-4 below, then follow **7b**
instead of step 5.
1. Copy `deploy\windows\web.config` to `APP_ROOT\web.config`. If `APP_ROOT` is not
`C:\shopdb-flask`, fix the paths inside it. Create `APP_ROOT\logs`.
2. Create an app pool with **No Managed Code**:
```powershell
Import-Module WebAdministration
New-WebAppPool -Name shopdbflask
Set-ItemProperty IIS:\AppPools\shopdbflask -Name managedRuntimeVersion -Value ""
```
3. Grant the app-pool identity access:
```powershell
icacls APP_ROOT /grant "IIS AppPool\shopdbflask:(OI)(CI)RX" /T
icacls APP_ROOT\logs /grant "IIS AppPool\shopdbflask:(OI)(CI)M" /T
mkdir APP_ROOT\instance 2>NUL
icacls APP_ROOT\instance /grant "IIS AppPool\shopdbflask:(OI)(CI)M" /T
```
4. **Unlock the handler sections** (locked server-wide by default; without this
IIS returns **HTTP 500.19**):
```powershell
%windir%\system32\inetsrv\appcmd unlock config /section:system.webServer/handlers
%windir%\system32\inetsrv\appcmd unlock config /section:system.webServer/httpPlatform
```
5. Create the site (own port; the classic ASP site can keep 8080):
```powershell
New-Website -Name shopdb-flask -Port 8090 -PhysicalPath APP_ROOT -ApplicationPool shopdbflask
New-NetFirewallRule -DisplayName "shopdb-flask 8090" -Direction Inbound -Protocol TCP -LocalPort 8090 -Action Allow
Start-Website shopdb-flask
```
IIS launches `waitress-serve --port=%HTTP_PLATFORM_PORT% wsgi:app` per the
web.config and reverse-proxies the site port to it. First request takes ~15s
(the app boots + connects to MySQL).
### 7b. Method B: subpath under an existing site
The mount path must match in **three places**: the IIS Application alias, the
`MOUNT_PATH` the backend sees, and the `VITE_BASE_PATH` the frontend was built
with. `/ops` is the example throughout; any alias works.
1. Rebuild the frontend for the subpath (on the dev box, then copy `dist`):
```bash
cd frontend && VITE_BASE_PATH=/ops/ npm run build # note the trailing slash
```
2. Create the Application under the existing site (instead of `New-Website`):
```powershell
New-WebApplication -Site "Default Web Site" -Name ops -PhysicalPath APP_ROOT -ApplicationPool shopdbflask
```
3. Tell the backend its mount path: in `APP_ROOT\web.config`, uncomment the
`MOUNT_PATH` environment variable (value `/ops`), or set `MOUNT_PATH=/ops`
in `APP_ROOT\.env`. `wsgi.py` then serves everything under the prefix
(requests outside it get a plain 404 naming the mount).
4. Recycle the app pool. The app is at `http(s)://<host>/ops/` and the API at
`/ops/api/...`.
The handler mappings in the app's web.config apply only inside the
Application, so the parent site's own handlers (classic ASP, static files)
are untouched. `CORS_ORIGINS` in `.env` is origin-only (scheme + host + port,
no path), so it is the same for both methods.
> The `X-Forwarded-For` URL Rewrite rule in web.config is **commented out by
> default**. It needs the URL Rewrite module; with it active but the module
> absent, IIS returns 500.19. Install URL Rewrite, then uncomment the
> `<rewrite>` block, to record real client IPs in audit logs.
>
> Two companion requirements, or the app keeps seeing 127.0.0.1:
> `allowedServerVariables` is locked at server level by default (500.52 when
> the block activates) - unlock once with
> `appcmd unlock config -section:system.webServer/rewrite/allowedServerVariables`.
> And waitress 2+ strips X-Forwarded-For from untrusted proxies, so the
> waitress `arguments` line must carry
> `--trusted-proxy=127.0.0.1 --trusted-proxy-headers=x-forwarded-for`
> (the shipped web.config already does).
---
## 8. Smoke test + first run
```powershell
(Invoke-WebRequest http://localhost:8090/ -UseBasicParsing).StatusCode # 200 (SPA)
Invoke-WebRequest http://localhost:8090/api/auth/login -Method POST `
-Body '{"username":"admin","password":"<the printed password>"}' `
-ContentType application/json -UseBasicParsing # 200 + token
```
Browse to `http://<host>:8090`, sign in as the admin, and the **setup wizard**
walks through site name, features (per-plugin: create tables here vs connect a DB),
floor-map upload, and starter data. Multiple Flask apps can share one IIS box -
each gets its own site, app pool, port, and venv.
---
## Troubleshooting
| Symptom | Cause / fix |
| --- | --- |
| `flask db upgrade` -> error **1071** | MySQL 5.6 without the step-1 flags (or server not restarted). |
| IIS **500.19** | handler sections not unlocked (step 7.4), or the `<rewrite>` block active without URL Rewrite. |
| IIS **500.52** after enabling the rewrite block | `allowedServerVariables` locked at server level - `appcmd unlock config -section:system.webServer/rewrite/allowedServerVariables`. |
| Audit log shows only **127.0.0.1** with the rewrite block active | waitress strips untrusted proxy headers - `--trusted-proxy=127.0.0.1 --trusted-proxy-headers=x-forwarded-for` missing from the waitress `arguments`. |
| **500** with an empty HttpPlatform log | app-pool identity can't read `APP_ROOT` / run the venv (step 7.3), or `.env` missing/invalid. |
| "internal error" toggling plugins, or uploads fail | app pool cannot WRITE `APP_ROOT\instance` (plugin registry, logos, photos, files live there) - step 7.3 grants it Modify. |
| "No time zone found with key America/New_York" | `tzdata` not installed (`pip install tzdata`). |
| Nav missing Machines/PCs/... | plugins not installed (step 6 `flask plugin install`), or site not recycled. |
| Method B: blank page / assets 404 under `/ops` | frontend `dist` built without `VITE_BASE_PATH=/ops/` (step 7b.1). |
| Method B: SPA loads but every API call 404s | `MOUNT_PATH` unset or not matching the Application alias (step 7b.3). |
| ConfigError on boot | a required `.env` var missing or left at a dev default. |