Excessive logging for Vimsvc.TaskManager tasks flooding hostd.log after upgrade to esxi 7

Products

• VMware vSphere ESXi

Issue/Introduction

provide a work around to decrease unnecessary logging after host upgrade to version 7

Symptoms:
- Excessive logging for  Vimsvc.TaskManager tasks flooding hostd.log 

- Logs from the InternalStatsCollector's queryLatestVmStats
calls made from vpxa to gather stats start flooding the logs as each call
for each VM's stats logs, etc.

- Log entries looks Like below :
cat hostd.{9..0} hostd.log | grep TaskManager | grep "Task Created" | grep InternalStatsCollector.queryLatestVmStats | tail -n 50
2021-12-08T09:59:51.931Z info hostd[2099844] [Originator@6876 sub=Vimsvc.TaskManager opID=3a0a5120 user=vpxuser] Task Created : haTask--vim.InternalStatsCollector.queryLatestVmStats-1632578
2021-12-08T09:59:51.932Z info hostd[2099853] [Originator@6876 sub=Vimsvc.TaskManager opID=3a0a5121 user=vpxuser] Task Created : haTask--vim.InternalStatsCollector.queryLatestVmStats-1632579

Environment

VMware vSphere ESXi 7.0

Cause

Back in time between 6.7 and 7.0 a change was made to log all task creations
and completions on info level, including the invisible tasks in release builds
(which was done only for debug builds before that). This was prompted by the
change that made all internal tasks invisible and in turn non-retainable so
that hostd does not crash with out-of-memory due to an accumulation of tasks
that needn't be retained, which in turn caused them to not be logged anymore.

 this change introduced a regression in the amount of logs generated.
For example, the logs from the InternalStatsCollector's queryLatestVmStats
calls made from vpxa to gather stats start flooding the logs as each call
for each VM's stats logs, etc.

Resolution

This issue is resolved in vSphere ESXi 7.0 U3f (build number 20036589).

Workaround:
setting vimsvc-logging tthe level to quiet and then to info stops the excess logging