From 80fb1dfc1a12df462e9a375761bba001c3457fc3 Mon Sep 17 00:00:00 2001
From: Raymond Jacobson <ray@audius.co>
Date: Fri, 22 May 2026 16:43:29 -0700
Subject: [PATCH] fix(eth-indexer): make /eth/health O(1)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The previous GetHealth ran a UNION/COUNT across users +
associated_wallets to populate tracked_wallets, plus a COUNT(*) on
eth_wallet_balances for cached_wallets. On prod that's ~3.15M rows
through a seq scan + dedup sort and consistently times out (or hangs
the handler — there was no statement timeout). Cheap locally, lethal
in prod.

Drop both counts from the response. They were nice-to-have stats, not
liveness signals — a health endpoint that takes 30s to tell you the
indexer is alive is worse than no endpoint. If you need population
stats, query eth_wallet_balances directly.

What's left is all O(1):
- connected, rpc_configured, last_block_seen, last_event_at: in-memory
- checkpoint_block: single-row PK lookup on eth_indexer_checkpoints

Also add a 2s context timeout to the handler. Even if a future query is
added that turns slow, the request fails fast instead of hanging the
ingress.
---
 eth/indexer/eth_indexer.go | 25 +++++++------------------
 eth/indexer/server.go      | 11 ++++++++++-
 2 files changed, 17 insertions(+), 19 deletions(-)

diff --git a/eth/indexer/eth_indexer.go b/eth/indexer/eth_indexer.go
index 9f4e56ff..e33d8b26 100644
--- a/eth/indexer/eth_indexer.go
+++ b/eth/indexer/eth_indexer.go
@@ -501,30 +501,21 @@ type ethHealth struct {
 	LastBlockSeen   uint64     `json:"last_block_seen"`
 	CheckpointBlock uint64     `json:"checkpoint_block"`
 	LastEventAt     *time.Time `json:"last_event_at"`
-	TrackedWallets  int64      `json:"tracked_wallets"`
-	CachedWallets   int64      `json:"cached_wallets"`
 }
 
+// GetHealth returns indexer liveness in O(1) — all values are either in
+// memory or come from a single-row PK lookup. Wallet-population counts
+// previously lived on this response but were expensive on prod
+// (UNION/COUNT across users + associated_wallets ≈ 3M rows, no index, can
+// take 30s+) and don't actually answer "is the indexer alive?", which is
+// what a health endpoint is for. If you want population stats, query
+// eth_wallet_balances directly.
 func (e *EthIndexer) GetHealth(ctx context.Context, maxEventLagSecs int64) (*ethHealth, error) {
 	checkpoint, err := e.loadCheckpoint(ctx)
 	if err != nil {
 		return nil, fmt.Errorf("loading checkpoint: %w", err)
 	}
 
-	var tracked, cached int64
-	err = e.pool.QueryRow(ctx, `
-		SELECT
-			(SELECT COUNT(*) FROM (
-				SELECT LOWER(wallet) FROM users WHERE wallet IS NOT NULL AND wallet <> ''
-				UNION
-				SELECT LOWER(wallet) FROM associated_wallets WHERE chain = 'eth' AND is_delete = FALSE
-			) t) AS tracked,
-			(SELECT COUNT(*) FROM eth_wallet_balances) AS cached
-	`).Scan(&tracked, &cached)
-	if err != nil {
-		return nil, fmt.Errorf("counting wallets: %w", err)
-	}
-
 	errs := make([]string, 0)
 	if !e.connected.Load() && e.wsURL != "" {
 		errs = append(errs, "websocket subscription not connected")
@@ -545,8 +536,6 @@ func (e *EthIndexer) GetHealth(ctx context.Context, maxEventLagSecs int64) (*eth
 		LastBlockSeen:   e.lastBlockSeen.Load(),
 		CheckpointBlock: checkpoint,
 		LastEventAt:     e.lastEventAt.Load(),
-		TrackedWallets:  tracked,
-		CachedWallets:   cached,
 	}, nil
 }
 
diff --git a/eth/indexer/server.go b/eth/indexer/server.go
index 82e0a55c..d5c68095 100644
--- a/eth/indexer/server.go
+++ b/eth/indexer/server.go
@@ -5,12 +5,19 @@ import (
 	"encoding/json"
 	"net"
 	"net/http"
+	"time"
 
 	"github.com/gofiber/fiber/v2"
 	"github.com/mcuadros/go-defaults"
 	"go.uber.org/zap"
 )
 
+// healthHandlerTimeout caps how long /eth/health is willing to wait on
+// downstream work (DB, etc.) before returning an error. The endpoint is
+// supposed to be O(1); a 2s ceiling is generous and stops a future slow
+// query from hanging the handler indefinitely.
+const healthHandlerTimeout = 2 * time.Second
+
 type Server struct {
 	*fiber.App
 	logger *zap.Logger
@@ -39,7 +46,9 @@ func NewServer(indexer *EthIndexer) *Server {
 		}
 		defaults.SetDefaults(&q)
 
-		health, err := indexer.GetHealth(c.Context(), q.MaxEventLagSecs)
+		ctx, cancel := context.WithTimeout(c.Context(), healthHandlerTimeout)
+		defer cancel()
+		health, err := indexer.GetHealth(ctx, q.MaxEventLagSecs)
 		if err != nil {
 			return c.Status(fiber.StatusInternalServerError).JSON(fiber.Map{
 				"error": err.Error(),