Two things of note:
First, processes can receive a signal that tells them that the system is going into hibernation. How they react to that is up to them but one possible action is that they terminate.
Second, network connections will most likely be broken when a hibernating instance is resumed because the remote end has timed out and not received any keepalives from the instance. This is probably going to be true for local processes that are communicating with each other via the network (even though the connections are internal) - the kernel will (again, probably) drop them due to timeout. This could cause the symptoms you're seeing.
I'm not sure this is a great answer but it's a possibility.
Can't launch ec2 spot instance with hibernation for instance type c5.18xlarasked 2 years ago
can I set up DNS to route requests to an alternate subdomain with EC2?asked 2 years ago
How does hibernation impact snapshots within EC2?Accepted Answerasked 7 months ago
Hibernation not working as expectedasked 7 months ago
Does AWS Batch support EC2 Spot stop/start and hibernationAccepted Answerasked 4 years ago
RDS Aurora does not start an instance that I had stopped, it has status starting and does not let me do anythingasked 6 months ago
EC2 instance stuck in Stopping stateasked a year ago
Trying to stop and terminate an instance, but it has a warning: "The instance 'i-0e571b9423e63be86' may not be stopped. Modify its 'disableApiStop' instance attribute and try again."asked a month ago
EC2 instance unresponsive and won't stopasked 4 years ago
My Website is Down and I Don't Know How to Get it Back Upasked 8 months ago