fulcrum: convert to a role, de-Uptime-Kuma the health check

685-line playbook becomes 33 lines plus a 426-line role
(install/service/healthcheck phases, six templates, one handler).
fulcrum_vars.yml is deleted; its content is the role's defaults.

Verified: fulcrum untouched - active since 2026-07-29 (no restart), 192G datadir,
height 966842, bitcoind and db_mem unchanged on disk. Second run changed=1 (the
arming run, changed_when: false aside). vipy changed=0.

THREE PRE-EXISTING LANDMINES the check-mode diff caught, any of which a faithful
extraction would have detonated:

- bitcoin_rpc_host was "192.168.1.140", commented "IP of knots_box_local". But
  .140 is fulcrum-box ITSELF; knots-box is .135. The DHCP leases had reshuffled -
  the fifth instance of this same disease in this estate. The live config had
  been hand-corrected to knots-box; running the playbook would have reverted it
  and pointed Fulcrum at itself. Now addressed by Tailscale name.

- The `Restart fulcrum` handler was guarded by uptime_kuma_enabled, so the three
  tasks that notify it (SSL cert, fulcrum.conf, systemd unit) could not restart
  anything. A config change applied to disk, reported success, and silently never
  took effect. That is worse than the other banner casualties: it makes the
  deployment itself lie. Ungated.

- db_mem was about to go 2048 -> 4448 (75% of 5931MB RAM), leaving ~1.4GB for the
  OS and Fulcrum's non-cache memory. The live value had been hand-tuned down.
  fulcrum_db_mem_mb_override pins it. Note set_fact outranks role defaults, so
  the calculation itself has to honour the override.

MY OWN ERROR, third instance: retyping `copy:` as `template:` lost `owner:` on
the banner and on fulcrum.conf. Rather than keep catching these by eye, every
managed path's owner/group/mode is now compared against `git show HEAD:`
mechanically - 12/12 match.

The health check timer had not fired since 2026-02-17 while reporting `active`
and `enabled`. It is OnBootSec + OnUnitActiveSec with no OnCalendar: OnBootSec
elapses once, and OnUnitActiveSec needs the SERVICE to have run this boot to have
anything to schedule from. Restarting the timer does not supply that; running the
service does, so the role now runs the check once after enabling. Also dropped
`Requires=fulcrum.service` from the timer - on a timer that means "stop watching
when the watched thing stops".

Diagnostic note: NextElapseUSecRealtime is always empty for a monotonic timer, so
it reads as broken even when healthy. I misread it once and wrongly called the
timer dead. Use NextElapseUSecMonotonic or systemctl list-timers.

fulcrum_ssl_port and fulcrum_tailscale_hostname moved to services_config.yml -
the socket-proxy play on the edge host needs them and a role default cannot reach
a second play.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

ansible/roles/fulcrum/defaults/main.yml

@@ -0,0 +1,78 @@
+# Fulcrum Configuration Variables
+
+# Version - Pinned to specific release
+fulcrum_version: "2.1.0"  # Fulcrum version to install
+
+# Directories
+fulcrum_db_dir: /mnt/fulcrum_data/fulcrum_db  # Database directory (heavy data on special mount)
+fulcrum_config_dir: /etc/fulcrum  # Config file location (standard OS path)
+fulcrum_lib_dir: /var/lib/fulcrum  # Other data files (banner, etc.) on OS disk
+fulcrum_binary_path: /usr/local/bin/Fulcrum
+
+# Network - Bitcoin RPC connection
+# Bitcoin Knots is on a different host (knots_box_local)
+# Using RPC user/password authentication (credentials from infra_secrets.yml)
+# Addressed by Tailscale name, never a LAN IP. This was
+#   bitcoin_rpc_host: "192.168.1.140"  # IP of knots_box_local
+# but .140 is fulcrum-box ITSELF - knots-box is .135. The DHCP leases had
+# reshuffled (the same drift that transposed the inventory), so running this
+# playbook would have pointed Fulcrum at itself and broken indexing. The live
+# config had already been hand-corrected to knots-box; this makes the repo
+# agree with it.
+bitcoin_rpc_host: "knots-box"
+bitcoin_rpc_port: 8332  # Bitcoin Knots RPC port
+# Note: bitcoin_rpc_user and bitcoin_rpc_password are loaded from infra_secrets.yml
+
+# Network - Fulcrum server
+fulcrum_tcp_port: 50001
+# Shared with the socket-proxy play on the edge host, so it lives in
+# services_config.yml rather than only here.
+fulcrum_ssl_port: "{{ service_settings.fulcrum.ssl_port }}"
+# Binding address for Fulcrum TCP/SSL server:
+#   - "127.0.0.1" = localhost only (use when Caddy is on the same box)
+#   - "0.0.0.0" = all interfaces (use when Caddy is on a different box)
+#   - Specific IP = bind to specific network interface
+fulcrum_tcp_bind: "0.0.0.0"  # Default: localhost (change to "0.0.0.0" if Caddy is on different box)
+fulcrum_ssl_bind: "0.0.0.0"  # Binding address for SSL port
+# If Caddy is on a different box, set this to the IP address that Caddy will use to connect
+
+# SSL/TLS Configuration
+fulcrum_ssl_enabled: true
+fulcrum_ssl_cert_path: "{{ fulcrum_config_dir }}/fulcrum.crt"
+fulcrum_ssl_key_path: "{{ fulcrum_config_dir }}/fulcrum.key"
+fulcrum_ssl_cert_days: 3650  # 10 years validity for self-signed cert
+
+# Port forwarding configuration (for public access via VPS)
+fulcrum_tailscale_hostname: "{{ service_settings.fulcrum.tailscale_hostname }}"
+
+# Performance
+# db_mem will be calculated as 75% of available RAM automatically in playbook
+# db_mem is computed as this share of RAM unless fulcrum_db_mem_mb is set
+# explicitly. On a 5931 MB host 75% is 4448 MB, which leaves ~1.4 GB for the
+# OS and for Fulcrum's non-cache memory; the live config had been hand-tuned
+# down to 2048 and that setting is respected below.
+fulcrum_db_mem_percent: 0.75  # 75% of RAM for database cache
+
+# Configuration options
+fulcrum_anon_logs: true  # Anonymize client IPs and TxIDs in logs
+fulcrum_peering: false  # Disable peering with other Fulcrum servers
+fulcrum_zmq_allow_hashtx: true  # Allow ZMQ hashtx notifications
+
+# Service user
+fulcrum_user: fulcrum
+fulcrum_group: fulcrum
+
+
+# --- Health check -----------------------------------------------------------
+# Checks the Electrum TCP port and records the answer in its exit code, which
+# systemd keeps: `systemctl is-failed fulcrum-healthcheck.service`.
+#
+# WHERE TO REPORT HEALTH — the one place to plug in monitoring. Empty means
+# check, exit honestly, report nowhere. Any endpoint accepting an HTTP ping
+# works; nothing here is specific to a monitoring product.
+healthcheck_push_url: ""
+
+# Explicit db_mem in MB. When set it wins over fulcrum_db_mem_percent; empty
+# means compute from RAM. Set here because the live host had been hand-tuned to
+# 2048 and a silent jump to 4448 is not something a refactor should do.
+fulcrum_db_mem_mb_override: 2048