Time to time sentinel report in logs that sentinel service itself down

[garry-t]

Hi,
I need to get clarification when on master I see time to time such logs. What is condition for this or how it detects that service is down?

9586:X 01 May 2024 21:22:08.164 # +sdown sentinel <hash> 10.2.0.10 26380 @ production-master 10.2.0.11 6380
9586:X 01 May 2024 21:22:20.896 # -sdown sentinel <hash> 10.2.0.10 26380 @ production-master 10.2.0.11 6380