A little update that may assist. If I look at the high level view of "all monitored machines" the issue is not present. This appear to be employing a different mechanism, but it confirms the data is present.

Thank you for bringing it to our attention, we are currently looking into this bug.

The issue lies with influxdb and the changing of step. We are on it, but until this is fixed you could try using victoria metrics as the main monitoring method. To do this adding this line DEFAULT_MONITORING_METHOD = "telegraf-victoriametrics" to settings/settings.py is needed.