ZFS tab: landing tier checks — stale files, verify passes against the mover's ledger, verify schedules

A verify pass reads the pool's copy of every file the mover moved back against the SHA-256 the
mover took of the landing copy (its ledger, named in its status file): match, mismatch, edited
since, gone, waiting, unreadable, unhashed. 'New moves' checks what moved since the last pass,
'everything' the whole ledger. Passes run as 'hamlab landing-verify' in their own session — the
same command a schedule's line in /etc/cron.d/hamlab-landing runs — so a console restart does
not cut one short. Stale files: anything waiting on the landing branch over an hour, listed.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
Mitchell-Johnson
2026-10-25 23:16:55 -07:00
co-authored by Claude Opus 5.5
parent f518df239f
commit 829617c932
8 changed files with 1275 additions and 2 deletions
+216
View File
@@ -0,0 +1,216 @@
package api
// The ZFS tab's landing-tier checks: what has waited too long on a landing branch, and verify
// passes that read the pool's copy of every moved file back against the mover's ledger —
// run by hand or on a schedule written to /etc/cron.d/hamlab-landing, just as snapshot
// schedules are. A pass runs as its own process (`hamlab landing-verify`), the same one cron
// starts, so it survives the console being restarted or upgraded mid-pass.
import (
"context"
"encoding/json"
"io"
"log"
"net/http"
"os"
"os/exec"
"path/filepath"
"syscall"
"time"
"hamlab/internal/storage"
)
func (s *Server) syncVerifyCron(ctx context.Context, z zfsState) error {
exe, err := os.Executable()
if err != nil {
return err
}
exe, _ = filepath.EvalSymlinks(exe)
return s.storage.WriteVerifyCron(ctx, z.Verify, s.user, exe, s.cfg.Path())
}
// startVerify launches a pass in its own session: arg is "new", "all" or a schedule id.
func (s *Server) startVerify(arg string) error {
exe, err := os.Executable()
if err != nil {
return err
}
exe, _ = filepath.EvalSymlinks(exe)
dir := filepath.Join(s.cfg.DataDir(), "landing-verify")
if err := os.MkdirAll(dir, 0o700); err != nil {
return err
}
out, err := os.OpenFile(filepath.Join(dir, "last.log"), os.O_CREATE|os.O_TRUNC|os.O_WRONLY, 0o600)
if err != nil {
return err
}
cmd := exec.Command(exe, "-config", s.cfg.Path(), "landing-verify", arg)
cmd.Stdout, cmd.Stderr = out, out
cmd.SysProcAttr = &syscall.SysProcAttr{Setsid: true}
if err := cmd.Start(); err != nil {
out.Close()
return err
}
go func() { _ = cmd.Wait(); out.Close() }()
return nil
}
func (s *Server) landingOverview(w http.ResponseWriter, r *http.Request) {
dir := s.cfg.DataDir()
now := time.Now()
type tier struct {
storage.LandingTarget
Cursor int64 `json:"cursor"`
StaleCount int `json:"stale_count"`
StaleBytes int64 `json:"stale_bytes"`
Stale []storage.StaleFile `json:"stale"`
}
tiers := []tier{}
for _, t := range storage.LandingTargets() {
files, n, b := storage.StaleFiles(t.Src, 100, now)
if files == nil {
files = []storage.StaleFile{}
}
tiers = append(tiers, tier{LandingTarget: t, Cursor: storage.VerifyCursor(dir, t), StaleCount: n, StaleBytes: b, Stale: files})
}
z := s.zfsLoad()
pols := make([]map[string]any, 0, len(z.Verify))
for i := range z.Verify {
p := z.Verify[i]
expr, _ := p.CronExpr()
m := map[string]any{"policy": p, "cron": expr, "last": storage.LastVerify(dir, p.ID)}
if p.Enabled {
if n := p.NextRun(now); !n.IsZero() {
m["next"] = n
}
}
pols = append(pols, m)
}
hist := storage.VerifyHistory(dir)
if len(hist) > 10 {
hist = hist[:10]
}
jsonOK(w, 200, map[string]any{
"tiers": tiers, "stale_after_sec": int(storage.StaleAfter.Seconds()),
"running": storage.RunningVerify(dir), "history": hist, "policies": pols,
"cron_file": storage.VerifyCronFile,
})
}
func (s *Server) landingVerifyNow(w http.ResponseWriter, r *http.Request) {
var body struct {
Scope string `json:"scope"`
}
if err := json.NewDecoder(io.LimitReader(r.Body, 1024)).Decode(&body); err != nil || (body.Scope != "new" && body.Scope != "all") {
jsonErr(w, 400, "scope: new or all")
return
}
if storage.RunningVerify(s.cfg.DataDir()) != nil {
jsonErr(w, 409, "a verify pass is already running")
return
}
if len(storage.LandingTargets()) == 0 {
jsonErr(w, 409, "no landing mover is reporting on this machine, so there is nothing to verify")
return
}
if err := s.startVerify(body.Scope); err != nil {
jsonErr(w, 500, err.Error())
return
}
log.Printf("landing verify (%s) started by %s", body.Scope, clientIP(r))
jsonOK(w, 202, map[string]string{"started": body.Scope})
}
func (s *Server) landingPolicyPut(w http.ResponseWriter, r *http.Request) {
var p storage.VerifyPolicy
if err := json.NewDecoder(io.LimitReader(r.Body, 16<<10)).Decode(&p); err != nil {
jsonErr(w, 400, "bad request")
return
}
if err := p.Validate(); err != nil {
jsonErr(w, 400, err.Error())
return
}
z := s.zfsLoad()
found := false
for i := range z.Verify {
if p.ID != "" && z.Verify[i].ID == p.ID {
z.Verify[i], found = p, true
}
}
if !found {
if p.ID == "" {
id, err := randomID()
if err != nil {
jsonErr(w, 500, err.Error())
return
}
p.ID = id
}
z.Verify = append(z.Verify, p)
}
if err := s.zfsSave(z); err != nil {
jsonErr(w, 500, err.Error())
return
}
if err := s.syncVerifyCron(r.Context(), z); err != nil {
jsonErr(w, 500, "saved, but writing "+storage.VerifyCronFile+" failed: "+err.Error())
return
}
expr, _ := p.CronExpr()
log.Printf("landing verify schedule %q saved by %s (%q, %s)", p.Name, clientIP(r), expr, p.Scope)
jsonOK(w, 200, p)
}
func (s *Server) landingPolicyDelete(w http.ResponseWriter, r *http.Request) {
id := r.PathValue("id")
z := s.zfsLoad()
kept := z.Verify[:0]
found := false
for _, p := range z.Verify {
if p.ID == id {
found = true
continue
}
kept = append(kept, p)
}
if !found {
jsonErr(w, 404, "no such schedule")
return
}
z.Verify = kept
if err := s.zfsSave(z); err != nil {
jsonErr(w, 500, err.Error())
return
}
if err := s.syncVerifyCron(r.Context(), z); err != nil {
jsonErr(w, 500, "removed, but writing "+storage.VerifyCronFile+" failed: "+err.Error())
return
}
storage.ForgetVerify(s.cfg.DataDir(), id)
log.Printf("landing verify schedule %s removed by %s", id, clientIP(r))
w.WriteHeader(204)
}
func (s *Server) landingPolicyRun(w http.ResponseWriter, r *http.Request) {
id := r.PathValue("id")
found := false
for _, p := range s.zfsLoad().Verify {
found = found || p.ID == id
}
if !found {
jsonErr(w, 404, "no such schedule")
return
}
if storage.RunningVerify(s.cfg.DataDir()) != nil {
jsonErr(w, 409, "a verify pass is already running")
return
}
if err := s.startVerify(id); err != nil {
jsonErr(w, 500, err.Error())
return
}
log.Printf("landing verify schedule %s run by %s", id, clientIP(r))
jsonOK(w, 202, map[string]string{"started": id})
}