Detailed steps to reproduce the problem?

Include full traceback, if possible

1. Increase logging on the server by running db.setProfilingLevel(0,-1) to ensure that monitoring commands are logged
2. Start a MongoClient and connect
3. Check the server logs for the monitoring thread's commands. The monitoring thread sends an initial isMaster and then switches to hello. The RTT however keeps flipping between ismaster and hello

I'm attaching server.log that shows the commands for the RTT thread and driver.log that shows some log lines I added to the RTT monitor to show hello_ok before and after the ping runs.

Definition of done: what must be done to consider the task complete?

hello_ok should be latched for the connection and the RTT thread should continue to use hello if isMaster returns helloOk

The exact Python version used, with patch level:

$ python -c "import sys; print(sys.version)"

3.11.4 (main, Apr 17 2026, 09:12:52) [Clang 17.0.0 (clang-1700.0.13.5)]

The exact version of PyMongo used, with patch level:

$ python -c "import pymongo; print(pymongo.version); print(pymongo.has_c())"

4.17.0
True

Describe how MongoDB is set up. Local vs Hosted, version, topology, load balanced, etc.

Local reproduction on my Mac. Set up a three node v8.0.19 replica set using m.

The operating system and version (e.g. Windows 7, OSX 10.8, ...)

ProductName: macOS
ProductVersion: 26.5.1
BuildVersion: 25F80

Web framework or asynchronous network library used, if any, with version (e.g. Django 1.7, mod_wsgi 4.3.0, gevent 1.0.1, Tornado 4.0.2, ...)

N/A

Security Vulnerabilities

If you’ve identified a security vulnerability in a driver or any other MongoDB project, please report it according to the instructions here