MONITOR Command

The MONITOR command streams every command processed by the Valkey server back to the client in real time. It is intended for debugging and development only.

MONITOR is supported in Python, Node.js, Java, and Go clients. C# and PHP support is not yet available.

How It Works

When you create a MonitorClient, GLIDE opens a dedicated connection and sends the MONITOR command. The server begins streaming command records back over that connection. Each record contains the timestamp, database index, client address, command name, and arguments.

MONITOR is only supported for standalone (non-cluster) connections. Attempting to use it with a cluster configuration will result in an error.

Caution

Running MONITOR in production degrades server throughput because every processed command must be serialized and sent back to the monitoring client. Use it only for short debugging sessions.

There are two modes for consuming monitor messages:

• Callback mode — provide a callback function at creation time; it is invoked for every incoming message.
• Queue mode — omit the callback; messages are queued internally and retrieved by calling getMonitorMessage / tryGetMonitorMessage.

Examples

Callback mode

Python

import asyncio
from glide import GlideClientConfiguration, NodeAddress
from glide.monitor_client import MonitorClient, MonitorMsg

def on_monitor_msg(msg: MonitorMsg) -> None:
    print(f"{msg.timestamp:.6f} [{msg.db}] {msg.client_addr} {msg.command} {msg.args}")

async def main():
    config = GlideClientConfiguration([NodeAddress("localhost", 6379)])
    async with await MonitorClient.create(config, callback=on_monitor_msg) as monitor:
        # Run commands on a separate client — the callback will print each one.
        await asyncio.sleep(5)

asyncio.run(main())

Node

import { GlideMonitorClient, MonitorLine } from "@valkey/valkey-glide";

const monitor = await GlideMonitorClient.create(
  { addresses: [{ host: "localhost", port: 6379 }] },
  (line: MonitorLine) => {
    console.log(`${line.timestamp} [${line.db}] ${line.clientAddr} ${line.command} ${line.args.join(" ")}`);
  },
);

// Close when done.
await monitor.close();

Java

import glide.api.MonitorClient;
import glide.api.models.commands.MonitorMsg;
import glide.api.models.configuration.GlideClientConfiguration;
import glide.api.models.configuration.NodeAddress;

GlideClientConfiguration config = GlideClientConfiguration.builder()
    .address(NodeAddress.builder().host("localhost").port(6379).build())
    .build();

try (MonitorClient monitor = MonitorClient.create(config, (MonitorMsg msg) ->
        System.out.printf("%.6f [%d] %s %s %s%n",
            msg.timestamp(), msg.db(), msg.clientAddr(), msg.command(), msg.args()))) {
    // Run commands on a separate client — the callback will print each one.
    Thread.sleep(5000);
}

Go

package main

import (
    "fmt"
    "time"

    "github.com/valkey-io/valkey-glide/go/v2"
    "github.com/valkey-io/valkey-glide/go/v2/config"
)

func main() {
    cfg := config.NewClientConfiguration().
        WithAddress(&config.NodeAddress{Host: "localhost", Port: 6379})

    monitor, err := glide.NewMonitorClient(cfg, func(line glide.MonitorLine) {
        fmt.Printf("%.6f [%d] %s %s %v\n",
            line.Timestamp, line.DB, line.ClientAddr, line.Command, line.Args)
    })
    if err != nil {
        panic(err)
    }
    defer monitor.Close()

    time.Sleep(5 * time.Second)
}

Queue mode

Python

import asyncio
from glide import GlideClientConfiguration, NodeAddress
from glide.monitor_client import MonitorClient

async def main():
    config = GlideClientConfiguration([NodeAddress("localhost", 6379)])
    async with await MonitorClient.create(config) as monitor:
        msg = await monitor.get_monitor_message()
        print(f"{msg.command} {msg.args}")

asyncio.run(main())

Node

import { GlideMonitorClient } from "@valkey/valkey-glide";

const monitor = await GlideMonitorClient.create({
  addresses: [{ host: "localhost", port: 6379 }],
});

const line = await monitor.getNextMessage();
console.log(`${line.command} ${line.args.join(" ")}`);

await monitor.close();

Java

import glide.api.MonitorClient;
import glide.api.models.commands.MonitorMsg;
import glide.api.models.configuration.GlideClientConfiguration;
import glide.api.models.configuration.NodeAddress;

GlideClientConfiguration config = GlideClientConfiguration.builder()
    .address(NodeAddress.builder().host("localhost").port(6379).build())
    .build();

try (MonitorClient monitor = MonitorClient.create(config)) {
    MonitorMsg msg = monitor.getMonitorMessage();
    System.out.printf("%s %s%n", msg.command(), msg.args());
}

Go

package main

import (
    "fmt"

    "github.com/valkey-io/valkey-glide/go/v2"
    "github.com/valkey-io/valkey-glide/go/v2/config"
)

func main() {
    cfg := config.NewClientConfiguration().
        WithAddress(&config.NodeAddress{Host: "localhost", Port: 6379})

    monitor, err := glide.NewMonitorClient(cfg, nil)
    if err != nil {
        panic(err)
    }
    defer monitor.Close()

    line, err := monitor.GetMonitorMessage()
    if err != nil {
        panic(err)
    }
    fmt.Printf("%s %v\n", line.Command, line.Args)
}

Monitor Message Fields

Each monitor message contains the following fields:

Field	Type	Description
timestamp	float/double	Unix timestamp of when the command was processed
db	int	Database index the command ran against
clientAddr	string	Address of the client that issued the command
command	string	Command name
args	list/array	Command arguments