Get started
Take a fresh Fidwyn appliance from power-on to a licensed, trusted, signed-in console. Fidwyn ships as a pre-baked golden image, so there is no operating system to install and no interactive installer to click through. A first-boot identity engine does the per-appliance work, and the steps below are the operator tasks that remain. Follow them in order; each one is a prerequisite for the next.
- A hypervisor (VMware, Proxmox, Hyper-V, KVM) or a bare-metal x86-64 host. Evaluation sizing: 8 vCPU, 16 GB RAM, 100 GB disk. Size production nodes per the deployment guide.
- Optional: an NVIDIA GPU passed through to the appliance if you will use Augur's local AI. The appliance boots and runs fully without one, and only the AI features stay disabled.
- The appliance ISO from the Download Center, plus its published SHA-256 and GPG signature.
- A reachable static IPv4 address, subnet, gateway and DNS server for the appliance (DHCP is fine for an evaluation).
- A license, or the ability to request one during setup (step 6).
1 · Verify the image ¶
Check the ISO against the published checksum and GPG signature before you boot it. A mismatch means the media is wrong or tampered with, so stop and re-download.
# verify integrity and signature
sha256sum fidwyn-2.1.iso
gpg --verify fidwyn-2.1.iso.asc fidwyn-2.1.iso
2 · Create the VM and boot ¶
- Create the machine. UEFI firmware, the sizing above, and one NIC on the network you will manage the appliance from. Attach the verified ISO as the boot device.
- Power on. The disk image is already partitioned and installed, so there is no OS setup. The appliance boots straight to its first-boot console.
3 · Assign the network address ¶
At the first-boot console, set the appliance's static IP, subnet, gateway and DNS (or accept DHCP for a lab). This is the address you and your users will browse to, so give it a DNS A record if you can, since the certificate you install in step 8 should match that name.
4 · Let first-boot identity run ¶
On first power-on the identity engine does the per-appliance work automatically: it regenerates per-appliance certificates and secrets (no two appliances share a key), unseals the vault, and brings every service up healthy. This is the sub-minute step and needs no input.
5 · First sign-in ¶
Browse to https://<appliance-ip>/ and sign in with the first-boot administrator credentials shown on the console. You are prompted to change the password immediately. This is the break-glass administrator, so store the new password in your own vault. You now have a working console; the last three steps make it production-ready.
6 · Activate the license ¶
Open System Settings → License. Because Fidwyn runs offline, activation is a two-step exchange rather than a call home:
- Generate the request key. The console produces a request key bound to this appliance. Copy it.
- Get it signed. Send the request key to Fidwyn; you receive a signed license file in return.
- Activate. Paste the signed license back into License and apply. The tier, node count and feature flags (EPM, Augur) update immediately.
7 · Set the date and time ¶
Under System Settings → Server Settings, set the timezone and an NTP source (or the time manually on an isolated network). Do this before you install certificates or bind a directory: token lifetimes, audit timestamps, session-recording clocks and certificate validity windows all depend on a correct clock.
8 · Install the server certificate ¶
First-boot generated a self-signed certificate so the console is reachable immediately. For production, replace it under System Settings → Server Settings with a certificate from your own CA, issued to the DNS name from step 3, covering the web console and the session broker. Nothing about Fidwyn depends on a public certificate authority.
What to do next ¶
Your appliance is now licensed, on the network, correctly clocked and trusted. Continue with the operational setup, in this order:
- Administration: connect a directory (AD/LDAP), create tenants, add users and authorize them with access groups, wire up email, then onboard your first systems and vault their credentials.
- Clustering & HA: if you are deploying more than one node, enroll the first node now and join the rest with a token.
- Hardening: air-gap, SIEM forwarding and backups before you go live.
