Conectarse a una instancia de Memorystore for Redis Cluster
Organízate con las colecciones
Guarda y clasifica el contenido según tus preferencias.
En esta página se explica cómo conectarse a una instancia de Memorystore for Redis Cluster.
Asistencia al cliente
Memorystore for Redis Cluster no proporciona sus propias bibliotecas de cliente, sino que admite bibliotecas de cliente de terceros que admiten el protocolo Redis.
Puedes consultar una lista de clientes de ejemplo en la página Clientes de Redis.
Práctica recomendada para el cliente redis-py
Para conectarte a tu instancia de Memorystore for Redis Cluster mediante el cliente de Python redis-py, debes añadir skip_full_coverage_check=True al declarar un RedisCluster:
Incluir puertos en la lista de permitidos del cortafuegos
Si no añades los puertos correctos a la lista de permitidos de tu cortafuegos, tu instancia puede tener errores de conexión. Para obtener más información sobre cómo añadir puertos a la lista de permitidos, consulta Solucionar problemas.
Acceso entre regiones
Puedes acceder a una instancia de Memorystore for Redis Cluster desde un cliente que se encuentre en una región distinta a la de la instancia de Memorystore.
En la sección Conectarse a esta instancia, anota la dirección IP y el número de puerto que aparecen junto a Endpoint de descubrimiento. Estos dos valores se usan juntos para formar el endpoint de detección de tu clúster.
gcloud
Para ver el endpoint de descubrimiento de tu clúster, ejecuta la describe
herramienta de línea de comandos gcloud como se explica en la sección Ver detalles de la instancia.
Anota la dirección del endpoint y el número de puerto que se indican en la sección discovery_endpoints:. Estos dos valores combinados son el endpoint de descubrimiento de tu clúster.
Conectarse desde una VM de Compute Engine mediante redis-cli
Puedes conectarte a la instancia de Memorystore for Redis Cluster desde cualquier VM de Compute Engine que use la red autorizada de la instancia de Memorystore.
Para conectarte a la instancia, sigue estos pasos:
Si aún no tienes una máquina virtual de Linux de Compute Engine que use la red autorizada de tu instancia de clúster de Memorystore para Redis, crea una y conéctate a ella siguiendo la guía de inicio rápido para crear una máquina virtual de Linux.
Instala redis-cli versión 6.0 o posterior en la VM de Compute Engine siguiendo las instrucciones de Instalar Redis en Linux.
DISCOVERY_ENDPOINT_ADDRESS y PORT_NUMBER son los valores que has anotado en el paso anterior.
.
Ejecuta el comando CLUSTER SHARDS para ver la topología de tu clúster. Anota una de las direcciones IP y los números de puerto del nodo.
Conéctate al nodo que quieras ejecutando el siguiente comando:
redis-cli -h NODE_IP_ADDRESS -p NODE_PORT -c
Haz los cambios siguientes:
NODE_IP_ADDRESS es la dirección IP del nodo que has encontrado en el paso anterior.
NODE_PORT es el número de puerto del nodo que has encontrado en el paso anterior.
Una vez que te hayas conectado al nodo, introduce algunos comandos de Redis:
Introduce estos datos:
PING
Resultado:
PONG
Intro
SET HELLO WORLD
Resultado:
OK
Introduce estos datos:
GET HELLO
Resultado:
"WORLD"
Cuando hayas terminado de probar la conexión al nodo de Redis, te recomendamos que elimines la VM de Compute Engine que has usado para conectarte a la instancia de Redis. De esta forma, evitarás que se te apliquen cargos en tu cuenta de facturación de Cloud.
Conectarse a una instancia con el cifrado en tránsito habilitado
En esta sección se muestra un ejemplo de cómo conectarse a una instancia de Memorystore for Redis Cluster que tiene habilitado el cifrado en tránsito.
Si aún no tienes una máquina virtual de Compute Engine que use la misma red autorizada que la instancia de Redis del paso anterior, crea una y conéctate a ella siguiendo la guía de inicio rápido con una máquina virtual de Linux.
Instala la versión redis-cli 6.0 o posterior en la VM de Compute Engine siguiendo las instrucciones de Instalar Redis en Linux.
[[["Es fácil de entender","easyToUnderstand","thumb-up"],["Me ofreció una solución al problema","solvedMyProblem","thumb-up"],["Otro","otherUp","thumb-up"]],[["Es difícil de entender","hardToUnderstand","thumb-down"],["La información o el código de muestra no son correctos","incorrectInformationOrSampleCode","thumb-down"],["Me faltan las muestras o la información que necesito","missingTheInformationSamplesINeed","thumb-down"],["Problema de traducción","translationIssue","thumb-down"],["Otro","otherDown","thumb-down"]],["Última actualización: 2025-09-10 (UTC)."],[],[],null,["# Connect to a Memorystore for Redis Cluster instance\n\nThis page explains how to connect to a Memorystore for Redis Cluster instance.\n\nClient support\n--------------\n\nMemorystore for Redis Cluster does not provide its own client libraries, but rather\nsupports existing third-party client libraries that support the Redis protocol.\nAn example list of clients can be found on the Redis [Clients](https://redis.io/docs/clients/)\npage.\n\n### Redis-py client best practice\n\nTo connect to your Memorystore for Redis Cluster instance using the [redis-py](https://github.com/redis/) Python client, you must add the `skip_full_coverage_check=True` when declaring a Redis\nCluster: \n\n from rediscluster import RedisCluster\n endpoints = [{\"host\": \"IPADDRESS\", \"port\": \"6379\"}]\n\n rdb = RedisCluster(\n startup_nodes=endpoints,\n skip_full_coverage_check=True, # Required for Memorystore\n decode_responses = True)\n\n print(rdb.set('PYTHON', 'CLUSTER'))\n print(rdb.get('PYTHON'))\n print(rdb.unlink('PYTHON'))\n\nAllowlist ports on firewall\n---------------------------\n\nIf you don't allowlist the correct ports on your firewall, your instance can\nencounter connection errors. For more information about allowlisting ports, see\n[Troubleshoot issues](/memorystore/docs/cluster/troubleshoot-issues#connection_error_caused_by_firewall_rules).\n\nCross region access\n-------------------\n\nYou can access a Memorystore for Redis Cluster instance from a client located in a\ndifferent region than where the Memorystore instance is located.\n\nView your cluster's discovery endpoint\n--------------------------------------\n\nTo view your Memorystore cluster's [discovery endpoint](/memorystore/docs/cluster/overview#discovery_endpoint), use the following instructions: \n\n### Console\n\n1. Go to the **Memorystore for Redis Cluster** page in the Google Cloud console.\n\n [Memorystore for Redis Cluster](https://console.cloud.google.com/memorystore/redis/clusters?)\n2. Click your Cluster ID.\n\n3. In the **Connect to this instance** section, take a note of the IP address\n and port number next to **Discovery Endpoint**. These two values used\n together are your cluster's discovery endpoint.\n\n### gcloud\n\nTo view your cluster's discovery endpoint, run the `describe`\ngcloud CLI as explained in the [view instance details](/memorystore/docs/cluster/create-manage-instances#view_instance_details) section.\n\nThe output should be similar to the following: \n\n authorizationMode: AUTH_MODE_DISABLED\n createTime: '2023-08-04T17:49:07.988918793Z'\n discoveryEndpoints:\n - address: 10.142.0.13\n port: 6379\n pscConfig:\n - network: projects/my-project-364018/global/networks/default\n name: projects/my-project-364018/locations/us-east1/clusters/my-cluster-1\n pscConnections:\n - address: 10.142.0.13\n forwardingRule: https://www.googleapis.com/compute/v1/projects/my-project-364018/regions/us-east1/forwardingRules/sca-auto-fr-5bc46f69-ec80-4c5a-8971-ce267a4d8ddd\n network: projects/my-project-364018/global/networks/default\n projectId: my-project-364018\n pscConnectionId: '25734306048376845'\n - address: 10.142.0.10\n forwardingRule: https://www.googleapis.com/compute/v1/projects/my-project-364018/regions/us-east1/.\n forwardingRules/sca-auto-fr-817b64c5-85ac-48d0-82e3-c6fa99b4e67d\n network: projects/my-project-364018/global/networks/default\n projectId: my-project-364018\n pscConnectionId: '25734306048376842'\n replicaCount: 0\n shardCount: 3\n sizeGb: 39\n state: ACTIVE\n tier: PERFORMANCE\n transitEncryptionMode: TRANSIT_ENCRYPTION_MODE_DISABLED\n uid: cddda5de-d91f-449b-8755-3a665430271a\n\nTake a note of the endpoint address and port number listed in the\n`discovery_endpoints:` section. These two values used together are your\ncluster's discovery endpoint.\n\nConnect from a Compute Engine VM using redis-cli\n------------------------------------------------\n\nYou can connect to the Memorystore for Redis Cluster instance from any\nCompute Engine VM that uses the Memorystore instance's\nauthorized network.\n\nTo connect to the instance:\n\n1. If you don't already have a Linux Compute Engine VM that uses the\n authorized network for your Memorystore for Redis Cluster instance, create one and\n connect to it by following [Quickstart to create a Linux VM](/compute/docs/create-linux-vm-instance).\n\n2. Install `redis-cli` version 6.0 or newer on the Compute Engine VM by\n following the instructions at [Install Redis on Linux](https://redis.io/docs/getting-started/installation/install-redis-on-linux/).\n\n3. [View your cluster's discovery endpoint](/memorystore/docs/cluster/connect-cluster-instance#view_your_clusters_discovery_endpoint)\n and make a note of it.\n\n4. Connect to your instance's discovery endpoint by running the following\n command:\n\n ```\n redis-cli -h DISCOVERY_ENDPOINT_ADDRESS -p PORT_NUMBER -c\n ```\n\n Where:\n - \u003cvar translate=\"no\"\u003eDISCOVERY_ENDPOINT_ADDRESS\u003c/var\u003e and \u003cvar translate=\"no\"\u003ePORT_NUMBER\u003c/var\u003e are the values you noted in the previous step.\n\n | **Note:** The `-c` switch is used to enable the `redis-cli` cluster support mode. If the `-c` switch is not provided, when you try to access keys from a different shard than the one you are connected to, it results in a `MOVED` error. In the cluster support mode the client is automatically reconnected to the address returned in the error response. For more details, see [Using Redis-CLI with a Redis Cluster](https://developer.redis.com/operate/redis-at-scale/scalability/redis-cli-with-redis-cluster/).\n5. Run the `CLUSTER SHARDS` command to view your cluster topology. Take a note\n of one of the node's IP addresses and port numbers.\n\n6. Connect to your desired node by running the following command:\n\n ```\n redis-cli -h NODE_IP_ADDRESS -p NODE_PORT -c\n ```\n\n Replace the following:\n - \u003cvar translate=\"no\"\u003eNODE_IP_ADDRESS\u003c/var\u003e is the node's IP address you found in the previous step.\n - \u003cvar translate=\"no\"\u003eNODE_PORT\u003c/var\u003e is the node's port number you found in the previous step.\n7. Once connected to the node, enter some Redis commands:\n\n Enter: \n\n ```\n PING\n ```\n\n Result: \n\n ```\n PONG\n ```\n\n Enter \n\n ```\n SET HELLO WORLD\n ```\n\n Result: \n\n ```\n OK\n ```\n\n Enter: \n\n ```\n GET HELLO\n ```\n\n Result: \n\n ```\n \"WORLD\"\n ```\n8. Once you are done testing your connection to the Redis node, you should\n consider deleting the Compute Engine VM you used to connect to the\n Redis instance. Doing so helps you avoid incurring charges to your\n Cloud Billing account.\n\nConnect to an in-transit encryption enabled instance\n----------------------------------------------------\n\nThis section provides an example of how to connect to a Memorystore for Redis Cluster\ninstance that has in-transit encryption enabled.\n\n1. If you don't already have a Compute Engine VM that uses the same\n authorized network as the Redis instance from the previous step, create one\n and connect to it by following [Quickstart using a Linux VM](/compute/docs/create-linux-vm-instance).\n\n 1. Install `redis-cli` version 6.0 or newer on the Compute Engine VM by following the instructions at [Install Redis on Linux](https://redis.io/docs/getting-started/installation/install-redis-on-linux/).\n2. Install your instance's certificate authority on the Linux VM by following\n the instructions at [Install certificate authorities on your client](https://cloud.google.com/memorystore/docs/cluster/manage-in-transit-encryption#install_certificate_authorities_on_your_client).\n\n3. [View your cluster's discovery endpoint](/memorystore/docs/cluster/connect-cluster-instance#view_your_clusters_discovery_endpoint)\n and make a note of it.\n\n4. Connect to your instance's discovery endpoint by running the following\n command:\n\n ```\n redis-cli -h DISCOVERY_ENDPOINT_ADDRESS -p PORT_NUMBER -c --tls --cacert /tmp/server_ca.pem\n ```\n\n Where:\n - \u003cvar translate=\"no\"\u003eDISCOVERY_ENDPOINT_ADDRESS\u003c/var\u003e and \u003cvar translate=\"no\"\u003ePORT_NUMBER\u003c/var\u003e are the values you noted in the previous step.\n\n | **Note:** The `-c` switch is used to enable the `redis-cli` cluster support mode. If the `-c` switch is not provided, when you try to access keys from a different shard than the one you are connected to, it results in a `MOVED` error. In the cluster support mode the client is automatically reconnected to the address returned in the error response. For more details, see [Using Redis-CLI with a Redis Cluster](https://developer.redis.com/operate/redis-at-scale/scalability/redis-cli-with-redis-cluster/).\n5. Run the `CLUSTER SHARDS` command to view your cluster topology. Take a note\n of one of the node's IP addresses and port numbers.\n\n6. Connect to your desired node by running the following command:\n\n ```\n redis-cli -h NODE_IP_ADDRESS -p NODE_PORT -c --tls --cacert /tmp/server_ca.pem\n ```\n\n Replace the following:\n - \u003cvar translate=\"no\"\u003eNODE_IP_ADDRESS\u003c/var\u003e is the node's IP address you found in the previous step.\n - \u003cvar translate=\"no\"\u003eNODE_PORT\u003c/var\u003e is the node's port number you found in the previous step.\n7. Once connected to the node, enter some Redis commands:\n\n Enter: \n\n ```\n PING\n ```\n\n Result: \n\n ```\n PONG\n ```"]]
