Today I’ve discovered that we are currently unable to create a snaphot of the .kibana* indices.
We are using Elasticsearch 8.7.1 with SG FLX 1.3.0-es-8.7.1.
Following API call confirms that our cluster has multiple .kibana indices. Due to privacy concerns I won’t show the index names.
GET "https://elasticsearch:9200/_cat/indices/.kibana*?h=index&expand_wildcards=all" | wc -l 382
We’re using a NFS share as the snapshot repository:
GET _snapshot/elastic_snapshots
{
"elastic_snapshots" : {
"type" : "fs",
"uuid" : "xEgwX2ErSw6NgEK-kd2bXA",
"settings" : {
"location" : "/nfs/elastic_snapshots"
}
}
}
With following API call I want to create a snaphot of all .kibana* indices:
PUT "https://elasticsearch:9200/_snapshot/elastic_snapshots/kibana-backup?wait_for_completion=true&pretty" -H 'Content-Type: application/json' -d'
{
"indices": ".kibana*",
"ignore_unavailable": true,
"include_global_state": false
}
'
However, only the ‘.kibana-event-log*’ indices will get saved, no hints of the .kibana config and multi-tenant indices.
{
"snapshot" : {
"snapshot" : "kibana-backup",
"uuid" : "ONJTJFiEQI6JofIKoikaFw",
"repository" : "elastic_snapshots",
"version_id" : 8070199,
"version" : "8.7.1",
"indices" : [
".kibana-event-log-8.7.1-000003",
".kibana-event-log-7.10.2-000032",
".kibana-event-log-8.7.1-000002"
],
"data_streams" : [ ],
"include_global_state" : false,
"state" : "SUCCESS",
"start_time" : "2024-03-21T08:03:42.242Z",
"start_time_in_millis" : 1711008222242,
"end_time" : "2024-03-21T08:03:42.242Z",
"end_time_in_millis" : 1711008222242,
"duration_in_millis" : 0,
"failures" : [ ],
"shards" : {
"total" : 3,
"failed" : 0,
"successful" : 3
},
"feature_states" : [ ]
}
We were creating snapshots of the kibana indices before and I’m pretty sure it was working with ES7.
Unfortunately I cant’ say since what Elastic or SG version this got broken.
This is quite unfortunate since we have no other possibility to easily backup the Kibana data.
Regards,
Alex