Hello there, I'm having issues restoring an elasticsearch snapshot.
I've tried:
POST _snapshot/snapshot_repo/snap-EIHidJXeQWuHpnGfzR04Uw/_restore
{
"indices": "target_indicies"
}
but I've got:
{
"error" : {
"root_cause" : [
{
"type" : "snapshot_restore_exception",
"reason" : "[snapshot_repo:snap-EIHidJXeQWuHpnGfzR04Uw] snapshot does not exist"
}
],
"type" : "snapshot_restore_exception",
"reason" : "[snapshot_repo:snap-EIHidJXeQWuHpnGfzR04Uw] snapshot does not exist"
},
"status" : 500
}
Yet the .dat file is on my directory.
The snapshot was created with this CURL:
FILE_TM=$(date +%Y%m%d)
curl -X PUT "http://$ELASTIC_URL/snapshot/myproject_repository/snapshot$FILE_TM?wait_for_completion=true"
searching for the snapshot's uuid in the repository bucket, I've found two files
named as such: snap-UUID.dat and meta-UUID.dat
those are the file that I'm trying to restore.
My issue here is that I must run locally an instance that is running as a stateful set on a gke instance that I can't yet access.
I do only have access to the bucket where the snapshots are stored and to the log of the job that runs the snapshot.
Therefore: running
GET /_cat/snapshots/snapshot_repo?v
will not return the snapshot I'm trying to restore.
Is there a way to restore an instance based only to its indexes and snapshots?
Along with the .dat files (meta-UUID.dat and snap-UUID.dat) I do have access to those files:
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.