I guess, the usual "turning it off and on again" fix worked as it always does...
I disabled the geoip processor:
curl -k -X PUT "https://localhost:9200/_cluster/settings?pretty" -H 'Content-Type: application/json' -d'
{
"persistent" : {
"ingest.geoip.downloader.enabled" : false
}
}
'
Then stopped and started each node, waiting for the cluster to recover every time.
Then I set ingest.geoip.downloader.enabled to null (making it default to true) and the system logged:
[2022-02-25T18:24:39,462][INFO ][o.e.c.s.ClusterSettings ] [node-01] updating [ingest.geoip.downloader.enabled] from [false] to [true] │
[2022-02-25T18:24:41,693][INFO ][o.e.i.g.DatabaseNodeService] [node-01] successfully loaded geoip database file [GeoLite2-ASN.mmdb] │
[2022-02-25T18:24:44,967][INFO ][o.e.i.g.DatabaseNodeService] [node-01] successfully loaded geoip database file [GeoLite2-Country.mmdb] │
[2022-02-25T18:24:45,052][INFO ][o.e.i.g.DatabaseNodeService] [node-01] successfully loaded geoip database file [GeoLite2-City.mmdb] │
Thanks for your help, Stephen