Files
bootapi/internal/netbox/cache.go
T
unkinben 274c480b09
ci/woodpecker/pr/build Pipeline was successful
ci/woodpecker/pr/pre-commit Pipeline was successful
ci/woodpecker/pr/test Pipeline was successful
Initial bootapi: NetBox-driven PXE/kickstart boot service
bootapi replaces Cobbler's PXE/kickstart side. It resolves a PXE-booting host
from NetBox (by MAC or hostname), renders an iPXE boot script and a kickstart
from Go text/templates, and serves them over HTTP. The ENC half already moved to
encapi; this covers the provisioning/boot half.

What's here:
- cmd/bootapi + internal/{config,model,netbox,render,server}; embedded default
  templates under templates/ (AlmaLinux 9 + Fedora kickstarts, iPXE boot +
  unknown-MAC fallbacks) ported from Cobbler's boot/bootstrap contract.
- NetBox client (v4.x API) behind a Resolver interface with a short-TTL cache;
  tested against httptest fixtures using real NetBox JSON shapes.
- chi HTTP server: /ipxe/{mac}, /boot/ipxe?mac=, /ks/{ident}, healthz/readyz,
  Prometheus /metrics. Unknown MAC -> safe fallback iPXE (200), unknown KS -> 404.
- Secrets (root pw hash, ssh keys) injected at render time from env/Vault, never
  NetBox. Config is env-based per estate convention.
- Makefile (build/test/lint/docker + patch/minor/major), Dockerfile (distroless),
  .woodpecker (pre-commit, golangci-lint v2 + go test -race, docker build on PR;
  image push + Gitea binary release on v* tag), docs/ and example config.

go build/vet clean, go test -race green, golangci-lint v2 clean, pre-commit clean.

Claude-Session: https://claude.ai/code/session_015ur3i7D2azsMAWTSVABApv
2026-07-28 20:57:11 +10:00

90 lines
2.3 KiB
Go

package netbox
import (
"context"
"sync"
"sync/atomic"
"time"
"git.unkin.net/unkin/bootapi/internal/model"
)
// Cache wraps a Resolver with a short-TTL in-memory cache. PXE boots come in
// bursts (iPXE fetches the boot script, then the kickstart, then package repos
// hit repeatedly), so even a 30s TTL collapses many NetBox lookups per host
// while keeping the data fresh enough that a re-provisioned host picks up
// changes on its next boot.
type Cache struct {
inner Resolver
ttl time.Duration
now func() time.Time // injectable for tests
mu sync.Mutex
entries map[string]cacheEntry
hits atomic.Int64
misses atomic.Int64
}
// Hits returns the cumulative cache-hit count (published as a metric).
func (c *Cache) Hits() int64 { return c.hits.Load() }
// Misses returns the cumulative cache-miss count (published as a metric).
func (c *Cache) Misses() int64 { return c.misses.Load() }
type cacheEntry struct {
host *model.Host
exp time.Time
}
// NewCache wraps inner with a TTL cache. A non-positive ttl disables caching.
func NewCache(inner Resolver, ttl time.Duration) *Cache {
return &Cache{
inner: inner,
ttl: ttl,
now: time.Now,
entries: map[string]cacheEntry{},
}
}
// HostByMAC returns a cached host or resolves and caches one.
func (c *Cache) HostByMAC(ctx context.Context, mac string) (*model.Host, error) {
return c.lookup(ctx, "mac:"+normalizeMAC(mac), func() (*model.Host, error) {
return c.inner.HostByMAC(ctx, mac)
})
}
// HostByName returns a cached host or resolves and caches one.
func (c *Cache) HostByName(ctx context.Context, name string) (*model.Host, error) {
return c.lookup(ctx, "name:"+name, func() (*model.Host, error) {
return c.inner.HostByName(ctx, name)
})
}
func (c *Cache) lookup(_ context.Context, key string, resolve func() (*model.Host, error)) (*model.Host, error) {
if c.ttl <= 0 {
return resolve()
}
now := c.now()
c.mu.Lock()
if e, ok := c.entries[key]; ok && now.Before(e.exp) {
c.mu.Unlock()
c.hits.Add(1)
return e.host, nil
}
c.mu.Unlock()
c.misses.Add(1)
// Resolve outside the lock so a slow NetBox call doesn't block cache hits.
host, err := resolve()
if err != nil {
return nil, err
}
c.mu.Lock()
c.entries[key] = cacheEntry{host: host, exp: now.Add(c.ttl)}
c.mu.Unlock()
return host, nil
}