scieloorg/elasticsearch-snapshot repository overview

⁠elasticsearch-snapshot

This container can be used to easily manage Elasticsearch snapshots.

⁠Environment Variables

• ELASTICSEARCH_URL (default: http://elasticsearch-9200.service.consul:9200)
  • The URL to reach your Elasticsearch server.
  • NOTE: This must be the entire hostname or IP. Hostnames that expect a search domain to be added will not work.
• ESS_ABORT_IF_EMPTY (default: true)
  • This will abort the snapshot process if there are no indices on the server.
  • This is useful if you just bootstrapped a cluster, and want to restore before doing any backups. The repository will still be created, but no snapshots will be created until there are indices on the server.
• ESS_CREATE_IF_MISSING (defaut: false)
  • If this is true and the snapshot repository does not exist, then attempt to create it. See the Repository Auto-Creation⁠ section for more information.
• ESS_MAX_SNAPSHOTS (defaut: 0)
  • The maximum number of snapshots to allow. Set to 0 to disable entirely.
  • Note: Only snapshots with a scheduled- prefix will be counted and deleted. All other snapshots will not count against this limit.
• ESS_REPO_NAME required
  • The name of the repository to create snapshots under.
• ESS_WAIT_FOR_COMPLETION (default: true)
  • Whether the execution should wait until the snapshot has been created before exiting.

⁠Usage

This is best used with a timer and service combination in your cluster.

# elasticsearch-snapshot.timer
[Unit]
Description=Elasticsearch Scheduled Snapshot Timer

[Timer]
OnCalendar=hourly
Persistent=false

[Install]
WantedBy=multi-user.target

[X-Fleet]
Conflicts=%p.timer
MachineOf=%p.service

# elasticsearch-snapshot.service
[Unit]
Description=Elasticsearch Scheduled Snapshot
Requires=docker.service
After=docker.service

[Service]
TimeoutStartSec=0
Type=simple

ExecStartPre=-/usr/bin/docker pull webdestroya/elasticsearch-snapshot

ExecStart=/usr/bin/docker run --rm=true \
  -e ELASTICSEARCH_URL=http://elasticsearch-9200.service.consul:9200 \
  -e ESS_REPO_NAME=s3_repository \
  -e ESS_MAX_SNAPSHOTS=100 \
  -e ESS_WAIT_FOR_COMPLETION=true \
  scieloorg/elasticsearch-snapshot

[Install]
WantedBy=multi-user.target

[X-Fleet]
Conflicts=%p.service