The exact same issue noted in this systemd-devel list is happening.
When I try to restart services such as:
systemctl restart mysql
Failed to restart mysql.service: Unit var.mount is masked.
systemctl restart ypbind
Failed to restart ypbind.service: Unit tmp.mount is masked.
When I try to unmask tmp.mount it always returns the same:
systemctl status tmp.mount
● tmp.mount - /tmp
Loaded: masked (Reason: Unit tmp.mount is masked.)
Active: active (mounted) since Wed 2020-03-18 13:40:03 EDT; 2 weeks 5 days ago
Where: /tmp
What: tmpfs
Tasks: 0 (limit: 57833)
Memory: 56.0K
CPU: 1ms
CGroup: /system.slice/tmp.mount
Warning: Journal has been rotated since unit was started. Log output is incomplete or unavailable.
Then I saw a lot of RPC: Timed out from the NIS primary and slave servers.
Other errors:
Apr 6 21:06:25 ourserver kernel: rpc-srv/tcp: nfsd: got error -32 when sending 24 bytes - shutting down socket
Apr 6 21:06:25 ourserver automount[3945]: set_tsd_user_vars: failed to get passwd info from getpwuid_r
I do see a lot of cockpit errors such as the below but there are probably a symptom but not related to the cause:
Apr 6 20:25:41 ourserver cockpit-ws[2749919]: received unsupported HTTP method
Apr 6 20:25:41 ourserver cockpit-ws[2749919]: received invalid HTTP request line
Apr 6 20:25:41 ourserver cockpit-ws[2749919]: received invalid HTTP request line
Apr 6 20:25:44 ourserver cockpit-tls[2736293]: cockpit-tls: gnutls_handshake failed: The TLS connection was non-properly terminated.
Apr 6 20:25:44 ourserver cockpit-tls[2736293]: cockpit-tls: gnutls_handshake failed: An unexpected TLS packet was received.
Apr 6 20:25:44 ourserver cockpit-tls[2736293]: cockpit-tls: gnutls_handshake failed: A connection with inappropriate fallback was at
tempted.
Apr 6 20:25:51 ourserver cockpit-tls[2736293]: cockpit-tls: gnutls_handshake failed: The TLS connection was non-properly terminated.
I uninstalled cockpit-ws for now.
And just like in the systemd-devel thread, all of the systemctl status *.mount are showing masked.
Is this something new that could break NIS?
Edit: I may have found a work-around, as suggested here, I set PrivateTmp=false in /usr/lib/systemd/system/ypbind.service and ran systemctl daemon-reload, and am waiting until there are less users logged in to restart ypbind.
I believe this could be a part of Features/ServicesPrivateTmp.