44 lines
2 KiB
Markdown
44 lines
2 KiB
Markdown
# verae-ops — install, run, maintain
|
||
|
||
How to stand up Verae Time × Zapier on **Docker**, **Proxmox**, **generic VMs**, or **dedicated hardware**, and how to **link dependent services**.
|
||
|
||
**Forgejo:** https://git.georgelambert.org/marchon/verae-ops
|
||
**Clone:** `ssh://git@git.georgelambert.org:2223/marchon/verae-ops.git`
|
||
**Catalog:** https://zapier.georgelambert.org/packages/verae-ops/README.pdf
|
||
**System map:** https://zapier.georgelambert.org/overview/README.pdf
|
||
|
||
This is the operations repo. Application code lives in the other independent git repositories.
|
||
|
||
## Reading order
|
||
|
||
1. [01-dependencies.md](01-dependencies.md) — what must exist, who talks to whom
|
||
2. [02-docker.md](02-docker.md) — Compose (NATS 3-node cluster + HTTPS edges)
|
||
3. [03-proxmox.md](03-proxmox.md) — LXC / QEMU VMs
|
||
4. [04-virtual-servers.md](04-virtual-servers.md) — cloud or hypervisor VMs
|
||
5. [05-dedicated-hardware.md](05-dedicated-hardware.md) — bare metal (NS1-style)
|
||
6. [06-linking-services.md](06-linking-services.md) — env vars, URLs, replica floors
|
||
7. [07-maintenance.md](07-maintenance.md) — upgrade, backup, fleet, NATS
|
||
|
||
Public PDFs of the same files: `https://zapier.georgelambert.org/packages/verae-ops/<name>.pdf`.
|
||
|
||
## Non-negotiables
|
||
|
||
| Rule | Why |
|
||
|------|-----|
|
||
| Zapier cloud → **HTTPS only** (zappier-edge) | Never NATS, never `api.veraetime.net` |
|
||
| NATS binds **loopback or a private docker/VM net** | Not on the public NIC |
|
||
| Tree-node **min 3**, pause does not count | Bulk Merkle lookups |
|
||
| Private keys stay on disk; git stores **paths** | `machines.json` `identityFile` |
|
||
| Operator console is **loopback** `:3850` | Not a public site |
|
||
|
||
## Quick lab (one machine)
|
||
|
||
```bash
|
||
git clone ssh://git@git.georgelambert.org:2223/marchon/verae-ops.git
|
||
cd verae-ops
|
||
docker compose up --build
|
||
```
|
||
|
||
Then: zappier http://127.0.0.1:3000/ middleware http://127.0.0.1:3100/health NATS monitoring http://127.0.0.1:8222/
|
||
|
||
Production layout is three NATS nodes + fleet-spread workers; see [02-docker.md](02-docker.md) and [05-dedicated-hardware.md](05-dedicated-hardware.md).
|