Restore and Backup Elasticsearch Indices
summary
Explains how Elasticsearch snapshots work and gives actionable steps: register a repository, prefer NFS shared filesystem mounted on every node (or use S3/Azure/GCS/HDFS where available), trigger snapshots via curl + snapshot API with options like partial, include_global_state, and index filters, adjust replica counts when restoring, and automate retention and selection with a Curator-style job. Shows examples of failing edge cases (partial corrupt shards, missing repository) and demonstrates automation with bareos for scheduled backups; the demo mostly works but lap-over issues are fixed in the linked GitHub.
AI-written summary · about
chapters (24)
- 0:00Intro
- 1:04Who am I
- 2:14Elastic Search
- 4:42Shards
- 6:19Where to backup
- 8:28Backup method
- 11:45Conclusion
- 12:27Repo
- 14:42NFS
- 16:03Building the base
- 16:45Announcing the repository
- 17:47Compression settings
- 19:04Repository options
- 20:47Working with snapshots
- 22:56Considerations for snapshots
- 23:24In Plex
- 27:51Why to restore
- 29:48Historical search
- 30:26Single shot
- 31:26GitHub
- 33:38Create Repository
- 37:34Verify Repository
- 39:28Snapshot
- 52:23Troubleshooting
related talks
46mMastering Snapshots in OpenSearch: From Basics To Advanced Strategi...
48mMastering OpenSearch Snapshots: From Basics To Advanced Strategies
28mCSI Volume Snapshots: On the Way to Faster and Better Backups
54mScale Your Elasticsearch Cluster
01h 01mElasticsearch Do's, Don'ts and Pro-Tips
26mCurrent state of MongoDB Backup options
01h 01mBest Practices With Elasticsearch
44mSurviving Elasticsearch