SCAN

Use SCAN to walk through the keys of the database incrementally, a batch at a time.

Each call takes a cursor and returns the next cursor together with a batch of keys. Start with cursor 0 and keep calling with the cursor from the previous reply until the server returns 0 again, which marks the end of the iteration. Because the work is split over many short calls, SCAN never blocks the server the way KEYS can on a large keyspace.

MATCH filters the returned keys with a glob-style pattern, COUNT hints at how much work each call should do (a hint about effort, not a page size, so batches vary in length), and TYPE limits the reply to keys of one type. Filtering is applied after a batch has been read, so a call can legitimately return no keys at all while the cursor is still non-zero: only the cursor tells you when the iteration is over.

The guarantee is that every key present for the whole iteration is returned at least once. Keys added or removed while the scan runs may or may not show up, and a key can be returned more than once, so make the processing of each key idempotent. HSCAN, SSCAN, and ZSCAN apply the same mechanism inside a single collection.

Syntax#

SCAN <cursor> [MATCH <pattern>] [COUNT <count>] [TYPE <type>]

Arguments#

Important points#

• This operation can inspect a large part of the database. Prefer cursor-based scans where possible and avoid unbounded use on hot paths.
• The cursor is opaque. Start with 0 and continue until the server returns cursor 0; a single iteration may return no elements.

Response#

The reply reports the result of the operation. Error replies have the same shape in RESP2 and RESP3 and are surfaced as exceptions by the SDKs below.

Examples#

TCP examples use the TLS REDIS_URL from the Upstash console. REST examples use UPSTASH_REDIS_REST_URL and UPSTASH_REDIS_REST_TOKEN.

SCAN 0

import { Redis } from "@upstash/redis";const redis = Redis.fromEnv();const [cursor, keys] = await redis.scan(0, { match: "*" });

from upstash_redis import Redisredis = Redis.from_env()result = redis.scan(0)print(result)

import Redis from "ioredis";const redis = new Redis(process.env.REDIS_URL!);const result = await redis.scan("0");console.log(result);

import { createClient } from "redis";const client = await createClient({ url: process.env.REDIS_URL })  .on("error", console.error)  .connect();const result = await client.scan("0");console.log(result);

import osimport redisclient = redis.from_url(os.environ["REDIS_URL"])result = client.scan(0)print(result)

package mainimport (    "context"    "fmt"    "os"    "github.com/redis/go-redis/v9")func main() {    opts, err := redis.ParseURL(os.Getenv("REDIS_URL"))    if err != nil {        panic(err)    }    client := redis.NewClient(opts)    result, _, err := client.Scan(context.Background(), 0, "*", 0).Result()    if err != nil {        panic(err)    }    fmt.Println(result)}

import java.net.URI;import redis.clients.jedis.Jedis;try (Jedis jedis = new Jedis(new URI(System.getenv("REDIS_URL")))) {  Object result = jedis.scan("0");  System.out.println(result);}

use redis::TypedCommands;fn main() -> redis::RedisResult<()> {    let url = std::env::var("REDIS_URL").expect("REDIS_URL is not set");    let client = redis::Client::open(url)?;    let mut connection = client.get_connection()?;    let iter: redis::Iter<String> = connection.scan()?;    for key in iter {        println!("{key}");    }    Ok(())}