Today all our 20+ servers stopped unexpectedly every 30-60 minutes.
In logs i see this errors:
[ERROR] Instance doc lost, stopping server. Check datetime settings
[WARNING] Removing instance doc
[ERROR] Failed to check subscription status
Today 25 October, Subscription until 25 November.
UPDATE New Event: [WARNING] Stopping duplicate instance, check date time sync
There’s likely a host with an incorrect date or time. All hosts use the time to validate the status of other hosts. Any host with an incorrect time will effect the entire cluster. There is likely also a network issue on the server.
Thank you!Will try to check the time
Checked all servers, found one who have +30 seconds time. After configuration NTPD on it, problem gone.
Thank you for help Zach!