_thread_keepalive spams to much with timer logs

[dolshevsk]

Checklist

• I have included information about relevant versions
• I have verified that the issue persists when using the master branch of Mode.

Steps to reproduce

Hey there, I'm building quite intense streaming application and my logs just spammed with Timer _thread_keepalive-AIOKafkaConsumerThread woke up too late when the abs(drift) value between 0.3-0.5 what from my perspective isn't a big problem and doesn't need too much reaction.

Expected behavior

It would be great if max_drift could be configured somehow so it can be set to 1.0 when the lag is inappropriate, or produce a log only when self.drifting threshold is reached.

Actual behavior

Currently there is no option to configure either the max_drift or self.drifting thresholds.

Full traceback

[2023-09-29 15:41:39,425] [7] [INFO] Timer _thread_keepalive-AIOKafkaConsumerThread woke up too late, with a drift of +0.3595487903803587 runtime=6.0498714447021484e-06 sleeptime=1.3595487903803587

Versions

• Python version: 3.11
• Mode version: 0.3.5
• Operating system: osx

[asaguaro1]

I am also having this issue. It consistently produces INFO logs that are notifying me of a < 1s drift with no other consequences that I can tell. Have you attempted any workarounds?

[wbarnha]

I agree, it is really a nuisance, let me see what I can do to give us the option to tell the application that drift is inevitable and that we don't care.