That second one is because your "max_wal_size" is too small. Increase it so your checkpoints do not happen more often than your your "checkpoint_timeout" setting.
That seems like a connection issue. Is the connection ovr a LAN? Or a VPN? There might have been some interruption, from a damaged cable to an isp problem
Regards,
Alvaro Aguayo Operations Manager Open Comb Systems E.I.R.L.
I faced some issues on postgres replication as it was stopped with these errors :-
LOG: unexpected EOF on client connection with an open transaction LOG: could not send data to client: Broken pipe FATAL: connection to client lost on DB Master