Loopback-only, edge tier, conservative checkpointing for slow disks. No network exposure means the built-in pair stays valid and TLS is unnecessary; enforcement defaults still apply so runaway statements and abandoned transactions cannot wedge the node while nobody is watching.
plomid-server \
--data /mnt/data \
--profile edge \
--max-connections 20 \
--statement-timeout 60s \
--idle-in-transaction-timeout 120sNotes: --profile edge also tightens checkpoint thresholds for slow storage; longer timeouts reflect unattended batch work, still bounded. Back the node up with the offline tools (see Backup and restore); derived caches (vector ANN structures) rebuild from rows on reopen.
Was this page helpful? Thanks — noted locally, nothing is sent anywhere.