Files
KamadoPool/api/cmd/kamado-api/main.go
T
satoshi df0dbf89e5 Harden block-recording pipeline: P0 reliability fixes
Closes the silent-failure modes between "ckpool logs a solve" and
"block correctly displayed":

* Difficulty estimate matched mempool.space — the projection now uses
  (inEpoch + 1) intervals so it converges on Bitcoin Core's eventual
  retarget formula at end-of-epoch instead of undershooting by ~0.05–
  0.10 % throughout.

* Tailer resumes mid-log on restart — persists (inode, offset) to kv
  every EOF + on shutdown, and replays the unread tail next time. Any
  solve line written while kamado-api was down would previously be
  invisible forever.

* Background reconcile loop (60 s) retries hash/reward enrichment for
  blocks the original RPC missed, so a transient bitcoind-index race no
  longer permanently leaves a block hashless.

* Reorg detection: same loop compares each recent stored hash against
  getblockhash(height); a mismatch stamps orphaned_at. UI renders these
  strikethrough with a red "orphaned" tag instead of showing illusory
  rewards forever.

* InsertBlock now reports whether a row was actually inserted; the
  caller WARN-logs duplicate-height ignores so a re-mined orphaned
  height can't disappear silently.

* Submit-attempt vs confirmed counters surface failed submissions:
  every "Possible/Submitting block solve" log line increments
  block_submit_attempts; "Solved and confirmed" increments
  block_submits_confirmed. A growing gap means bitcoind is rejecting
  our submissions — previously invisible.

* share_err patch refreshed against pinned ckpool source: added
  SE_NO_JOBID -> 21 and SE_WORKER_MISMATCH -> 24 mappings, kept
  SE_INVALID_NONCE2 in 20 (it's a malformed-input error, not low-diff).
  AxeOS users now see actionable Stratum codes instead of
  "unknown error".

UI gets new orphaned_at + block_submit_attempts/confirmed fields on
the snapshot type and a strikethrough-with-tag rendering for orphaned
blocks in BlocksTable.
2026-04-27 16:30:15 +03:00

140 lines
4.1 KiB
Go

// kamado-api is the middleware that sits between ckpool-solo and the
// Kamado dashboard. It polls ckpool's Unix socket API, calls bitcoind
// over JSON-RPC, and serves the merged state over REST (WebSocket push,
// SQLite persistence, ZMQ and log-tailer-based block detection land in
// a follow-up commit).
package main
import (
"context"
"errors"
"fmt"
"log/slog"
"net/http"
"os"
"os/signal"
"strconv"
"strings"
"syscall"
"time"
"github.com/kamadopool/kamado-api/internal/bitcoind"
"github.com/kamadopool/kamado-api/internal/ckpool"
"github.com/kamadopool/kamado-api/internal/config"
"github.com/kamadopool/kamado-api/internal/httpapi"
"github.com/kamadopool/kamado-api/internal/logmon"
"github.com/kamadopool/kamado-api/internal/state"
"github.com/kamadopool/kamado-api/internal/store"
"github.com/kamadopool/kamado-api/internal/zmqmon"
)
func main() {
log := slog.New(slog.NewJSONHandler(os.Stdout, &slog.HandlerOptions{Level: slog.LevelInfo}))
slog.SetDefault(log)
cfg, err := config.FromEnv()
if err != nil {
log.Error("config", "err", err)
os.Exit(1)
}
log.Info("kamado-api starting",
"listen", cfg.ListenAddr,
"sockdir", cfg.CKPoolSockDir,
"bitcoind", cfg.BitcoinRPCURL,
"poll_interval", cfg.PollInterval,
)
ck := ckpool.New(cfg.CKPoolSockDir)
rpc := bitcoind.NewRPC(cfg.BitcoinRPCURL, cfg.BitcoinRPCUser, cfg.BitcoinRPCPassword, cfg.BitcoinRPCTimeout)
// Persistent block history. Non-fatal if it can't be opened — the
// aggregator falls back to an in-memory ring so the pool keeps
// running even with a broken data volume.
var blockStore *store.BlockStore
if cfg.DBPath != "" {
if s, err := store.Open(cfg.DBPath); err != nil {
log.Warn("block store open failed, running without persistence", "path", cfg.DBPath, "err", err)
} else {
blockStore = s
defer blockStore.Close()
log.Info("block store opened", "path", cfg.DBPath)
}
}
agg := state.New(ck, rpc, cfg.PollInterval, log)
agg.Store = blockStore
agg.MempoolBaseURL = cfg.MempoolBaseURL
ctx, stop := signal.NotifyContext(context.Background(), syscall.SIGINT, syscall.SIGTERM)
defer stop()
api := httpapi.New(agg, log)
// Wire snapshot refreshes into the WebSocket hub so subscribers get
// real-time updates without polling.
agg.OnRefresh = api.Hub.Broadcast
// Optional: bitcoind hashblock ZMQ subscription for sub-second
// chain-tip refreshes. Empty endpoint disables it.
zmq := zmqmon.New(cfg.BitcoinZMQBlock, log)
go zmq.Run(ctx)
go agg.Run(ctx, zmq.Events)
// Background reconciliation: retry hash/reward enrichment for blocks
// the initial RPC lookup couldn't fetch, and detect chain reorgs by
// comparing recorded hashes against the canonical chain.
go agg.ReconcileBlocks(ctx)
// Tail the ckpool log for block-solve events (our own solves).
tailer := logmon.New(cfg.CKPoolLogFile, log)
if blockStore != nil {
const cursorKey = "logmon_cursor"
tailer.LoadCursor = func() (uint64, int64, bool) {
v, err := blockStore.GetKV(cursorKey)
if err != nil || v == "" {
return 0, 0, false
}
parts := strings.SplitN(v, ":", 2)
if len(parts) != 2 {
return 0, 0, false
}
ino, err1 := strconv.ParseUint(parts[0], 10, 64)
off, err2 := strconv.ParseInt(parts[1], 10, 64)
if err1 != nil || err2 != nil {
return 0, 0, false
}
return ino, off, true
}
tailer.SaveCursor = func(ino uint64, off int64) {
if err := blockStore.SetKV(cursorKey, fmt.Sprintf("%d:%d", ino, off)); err != nil {
log.Warn("logmon cursor persist failed", "err", err)
}
}
}
go tailer.Run(ctx)
go agg.IngestBlockEvents(ctx, tailer.Events)
go agg.IngestAttemptEvents(ctx, tailer.Attempts)
srv := &http.Server{
Addr: cfg.ListenAddr,
Handler: api.Handler(),
ReadHeaderTimeout: 5 * time.Second,
}
// Shutdown on ctx cancel
go func() {
<-ctx.Done()
log.Info("shutting down")
shutdownCtx, cancel := context.WithTimeout(context.Background(), 10*time.Second)
defer cancel()
_ = srv.Shutdown(shutdownCtx)
}()
log.Info("http listening", "addr", cfg.ListenAddr)
if err := srv.ListenAndServe(); err != nil && !errors.Is(err, http.ErrServerClosed) {
log.Error("http server", "err", err)
os.Exit(1)
}
}