Elasticseach Backup and Restore

Elastic Stack Elasticsearch
1

Hi,
I want to backup an elasticsearch-oss:7.2.0 cluster running in Kubernetes and restore it into a new one elasticsearch:7.2.1 running within ECK

I created a snapshot which looks like to be good (40GB of data)
curl /_snapshot/my_repo/snap1 --> SUCCESS after about 10mn

However, when I restore it into the new cluster, it seems that indices with large datasets are missing:

curl /_snapshot/my_repo/snap1/_restore -d'
{
  "ignore_unavailable": true,
  "include_global_state": true,
  "rename_pattern": "index_(.+)",
  "rename_replacement": "restored_index_$1"
}'

Running /_cat/indices on old then new cluster shows that some data are missing:

OLD CLUSTER: green open ind1  aBoy4sS6Tp6_a9Xu3zuuMw 1 1  75   359     1mb 558.8kb
NEW CLUSTER: green open ind1  Dt8NrsYAQRGEfqOa049jPQ 1 1  75  7460 868.5kb 434.2kb

OLD CLUSTER: green open ind2  22IYbVNBSNWmHHKHpoAcmQ 1 1 62592 17113 2gb 773.1mb
NEW CLUSTER: red   open ind2  WsbAoc6nSpylZWqnzXvI-A 1 1

Any idea how to debug this ?
Find any error message ?
Am I missing something to run such backup from my old DB to the new one ?

Thanks a lot for your help!

2

I progress on the issue.

Running /_cat/_shards/and/_cluster/allocation/explain` helped to figure out what's going on.

Apparently, the name of my VolumeClaimTemplate was wrong, as it must be 'elasticsearch-data'
https://www.elastic.co/guide/en/cloud-on-k8s/current/k8s-volume-claim-templates.html

3

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.