Manual Ping

`client.ping()`

Send a PINGREQ and await the PINGRESP. Returns the round-trip time in seconds:

rtt = await client.ping()
print(f"Broker RTT: {rtt * 1000:.1f} ms")

Timeout parameter

rtt = await client.ping(timeout=5.0)   # raises MQTTTimeoutError if no response in 5 s

Default timeout is 10 seconds. MQTTTimeoutError is raised if PINGRESP does not arrive within the timeout.

from zmqtt import MQTTTimeoutError

try:
    rtt = await client.ping(timeout=2.0)
except MQTTTimeoutError:
    print("Broker is not responding")

Keepalive vs manual ping

The library automatically sends PINGREQ packets on the keepalive interval configured at connection time (see keepalive parameter, default 60 s). This keeps the TCP connection alive and lets the broker detect dead clients.

client.ping() is for application-level use — measuring latency, confirming broker reachability before a critical operation, or implementing a health check. It is independent of the automatic keepalive ping loop.

Concurrent pings

Multiple concurrent ping() calls are safe. Each waiter registers independently and receives its own RTT measurement when the next PINGRESP arrives.