Add resiliency primitives: atomic writes, reconcile mutex, state backup

- Add storage.WriteFileAtomic (temp + rename) and storage.CopyFile helpers
- Convert all 8 production config writers to atomic writes: config/state.yaml,
  dnsmasq, nftables, domains, wireguard (config + secrets + peers + wg0.conf),
  tunnel/cloudflared
- Add sync.Mutex to Reconciler to serialize concurrent Reconcile() calls
  triggered by domain registration goroutines
- Add state.yaml backup (.bak) before every write; LoadState falls back to
  backup if primary is corrupted
- Reconciler refuses to use empty config on corruption (only on first run
  when no state file exists yet)
This commit is contained in:
2026-07-14 11:37:33 +00:00
parent 3172e56288
commit 88acd437a1
9 changed files with 110 additions and 27 deletions

View File

@@ -6,6 +6,8 @@ import (
"os"
"os/exec"
"strings"
"github.com/wild-cloud/wild-central/internal/storage"
)
const defaultRulesPath = "/etc/nftables.d/wild-cloud.nft"
@@ -144,7 +146,7 @@ func (m *Manager) WriteRules(content string) error {
return err
}
if err := os.WriteFile(m.rulesPath, []byte(content), 0644); err != nil {
if err := storage.WriteFileAtomic(m.rulesPath, []byte(content), 0644); err != nil {
return fmt.Errorf("writing rules file: %w", err)
}
@@ -171,7 +173,7 @@ func (m *Manager) WriteDisabledRules() error {
"# Managed by Wild Cloud Central API — do not edit manually\n\n" +
"table inet wild-cloud {}\n" +
"delete table inet wild-cloud\n"
if err := os.WriteFile(m.rulesPath, []byte(content), 0644); err != nil {
if err := storage.WriteFileAtomic(m.rulesPath, []byte(content), 0644); err != nil {
return fmt.Errorf("writing disabled rules file: %w", err)
}
slog.Info("nftables rules disabled", "component", "nftables", "path", m.rulesPath)