anchorage v1.0 initial tree

Greenfield Go multi-tenant IPFS Pinning Service wire-compatible with the
IPFS Pinning Services API spec. Paired 1:1 with Kubo over localhost RPC,
clustered via embedded NATS JetStream, Postgres source-of-truth with
RLS-enforced tenancy, Fiber + huma v2 for the HTTP surface, Authentik
OIDC for session login with kid-rotated HS256 JWT API tokens.

Feature-complete against the 22-milestone build plan, including the
ship-it v1.0 gap items:

  * admin CLIs: drain/uncordon, maintenance, mint-token, rotate-key,
    prune-denylist, rebalance --dry-run, cache-stats, cluster-presences
  * TTL leader election via NATS KV, fence tokens, JetStream dedup
  * rebalancer (plan/apply split), reconciler, requeue sweeper
  * ristretto caches with NATS-backed cross-node invalidation
    (placements live-nodes + token denylist)
  * maintenance watchdog for stuck cluster-pause flag
  * Prometheus /metrics with CIDR ACL, HTTP/pin/scheduler/cache gauges
  * rate limiting: session (10/min) + anonymous global (120/min)
  * integration tests: rebalance, refcount multi-org, RLS belt
  * goreleaser (tar + deb/rpm/apk + Alpine Docker) targeting Gitea

Stack: Cobra/Viper, Fiber v2 + huma v2, embedded NATS JetStream,
pgx/sqlc/golang-migrate, ristretto, TypeID, prometheus/client_golang,
testcontainers-go.
This commit is contained in:
57_Wolve committed 2026-04-16 18:13:36 -05:00
commit 12bf35caf8
112 files changed
+16884

No files matched your search

+48
View File
@@ -0,0 +1,48 @@
[Unit]
Description=anchorage IPFS Pinning Service
Documentation=https://git.anomalous.dev/alphacentri/anchorage
Wants=network-online.target
After=network.target network-online.target postgresql.service ipfs.service
[Service]
Type=simple
User=anchorage
Group=anchorage
WorkingDirectory=/etc/anchorage
LimitNOFILE=65536
ExecStart=/usr/bin/anchorage serve --config /etc/anchorage/anchorage.yaml
Restart=on-failure
RestartSec=5
StartLimitInterval=60
StartLimitBurst=3
# Security hardening — anchorage is a pure-Go HTTP/NATS daemon, no
# special capabilities required. The writable paths are the state dir
# (node.id, NATS JetStream data) and the log dir.
ProtectSystem=strict
ProtectHome=true
PrivateTmp=true
PrivateDevices=true
ProtectKernelTunables=true
ProtectKernelModules=true
ProtectControlGroups=true
RestrictSUIDSGID=true
LockPersonality=true
NoNewPrivileges=true
RestrictRealtime=true
RestrictNamespaces=true
MemoryDenyWriteExecute=true
SystemCallArchitectures=native
SystemCallFilter=@system-service
SystemCallFilter=~@privileged @resources @mount @debug
ReadWritePaths=/var/lib/anchorage /var/log/anchorage
# Allow binding to ports < 1024 (e.g. :443 behind a reverse proxy terminator).
AmbientCapabilities=CAP_NET_BIND_SERVICE
CapabilityBoundingSet=CAP_NET_BIND_SERVICE
[Install]
WantedBy=multi-user.target
+29
View File
@@ -0,0 +1,29 @@
#!/bin/sh
set -e
systemctl daemon-reload
# Enable the service but don't start on a fresh install — the operator
# still needs to write /etc/anchorage/anchorage.yaml (start from
# /etc/anchorage/anchorage.yaml.example) and provision the JWT signing
# key at auth.apiToken.signingKeyPath.
systemctl enable anchorage.service >/dev/null 2>&1 || true
# On upgrade ("2" from rpm, "configure" from deb) restart only if the
# service was previously running so an operator who has it intentionally
# stopped stays that way.
if [ "$1" = "configure" ] || [ "$1" = "2" ]; then
if systemctl is-active --quiet anchorage.service; then
systemctl restart anchorage.service >/dev/null 2>&1 || true
fi
fi
if [ ! -f /etc/anchorage/anchorage.yaml ]; then
echo ""
echo "anchorage: configuration not found at /etc/anchorage/anchorage.yaml"
echo " cp /etc/anchorage/anchorage.yaml.example /etc/anchorage/anchorage.yaml"
echo " edit to point at your Postgres + Authentik + Kubo RPC endpoint"
echo " generate a signing key, e.g.: openssl rand -base64 48 > /etc/anchorage/jwt.key"
echo " systemctl start anchorage"
echo ""
fi
+18
View File
@@ -0,0 +1,18 @@
#!/bin/sh
set -e
systemctl daemon-reload
# On purge (deb) or final erase (rpm) the operator is removing anchorage
# for good — clean up the system user. /var/lib/anchorage is left in
# place so the operator can archive NATS state and node.id first.
if [ "$1" = "purge" ] || [ "$1" = "0" ]; then
userdel anchorage 2>/dev/null || true
groupdel anchorage 2>/dev/null || true
echo ""
echo "anchorage: state retained at /var/lib/anchorage and /var/log/anchorage"
echo " remove manually if not needed:"
echo " rm -rf /var/lib/anchorage /var/log/anchorage /etc/anchorage"
echo ""
fi
+16
View File
@@ -0,0 +1,16 @@
#!/bin/sh
set -e
# Create the anchorage system user and group if missing.
getent group anchorage >/dev/null 2>&1 || groupadd --system anchorage
getent passwd anchorage >/dev/null 2>&1 || useradd --system --no-create-home --shell /usr/sbin/nologin -g anchorage anchorage
# Directories the service needs. nfpm also creates these, but repeating
# here covers the case where the operator wiped /var/lib/anchorage
# between installs.
mkdir -p /etc/anchorage \
/var/lib/anchorage \
/var/lib/anchorage/nats \
/var/log/anchorage
chown -R anchorage:anchorage /etc/anchorage /var/lib/anchorage /var/log/anchorage
chmod 0750 /var/lib/anchorage /var/lib/anchorage/nats /var/log/anchorage
+5
View File
@@ -0,0 +1,5 @@
#!/bin/sh
set -e
systemctl stop anchorage.service >/dev/null 2>&1 || true
systemctl disable anchorage.service >/dev/null 2>&1 || true