Was this page helpful?
Caution
You're viewing documentation for a previous version of Scylla Manager. Switch to the latest stable version.
Examples¶
Prerequisites¶
Backup locations (buckets) with backed-up data created by backup task.
Snapshot tag of backup that you want to restore.
To get a list of all available snapshot tags, use sctool backup list command.
Schedule restore¶
Use the example below to run the sctool restore command.
sctool restore -c <id|name> -L <list of locations> -T <Snapshot tag> --restore-<schema|tables>
restore/3208ff15-6e8f-48b2-875c-d3c73f545410
where:
-c- the name or ID of the restore destination cluster-L- points to backup storage location ins3:<your S3 bucket name>format-T- snapshot tag that you want to restore--restore-schema or --restore-tables- type of restore task
The command returns the task ID. You will need this ID for additional actions.
Complete the restore procedure - restore both the schema and the content of the tables¶
Ensure that all prerequisites of restoring schema are met.
The easiest way to achieve that is to perform restore schema on a new empty cluster.
Restore schema using sctool restore.
sctool restore -c prod-cluster -L s3:test-bucket -T snapshot_tag --restore-schema restore/3208ff15-6e8f-48b2-875c-d3c73f545410
Ensure that restoring schema has finished using sctool progress.
sctool progress restore/3208ff15-6e8f-48b2-875c-d3c73f545410 -c prod-cluster Run: 774ce0fe-a7b4-11ed-82da-0892040e83bb Status: DONE - restart required (see restore docs) Start time: 16 Feb 23 17:09:55 CET End time: 16 Feb 23 17:09:59 CET Duration: 4s Progress: 100% | 100% Snapshot Tag: sm_20230216160631UTC ╭───────────────┬─────────────┬──────────┬──────────┬────────────┬──────────────┬────────╮ │ Keyspace │ Progress │ Size │ Success │ Downloaded │ Deduplicated │ Failed │ ├───────────────┼─────────────┼──────────┼──────────┼────────────┼──────────────┼────────┤ │ system_schema │ 100% | 100% │ 557.728k │ 557.728k │ 557.728k │ 0 │ 0 │ ╰───────────────┴─────────────┴──────────┴──────────┴────────────┴──────────────┴────────╯
Perform restore schema follow-up action - rolling restart.
Ensure that all prerequisites of restoring the content of the tables are met.
The easiest way to achieve that is to restore schema with sctool restore (so that’s what has already been done in this example). You don’t need to follow this step if you are sure that the destination cluster has the correct schema of restored tables and that those tables are truncated.
Restore the content of the tables using sctool restore.
sctool restore -c prod-cluster -L s3:test-bucket -T snapshot_tag --restore-tables -K test_keyspace restore/31043443-8214-42d3-be98-ed9771c48dde
Ensure that restoring the content of the tables has finished using sctool progress.
sctool progress restore/31043443-8214-42d3-be98-ed9771c48dde -c prod-cluster --details Run: e10ec718-a7bf-11ed-b05a-0892040e83bb Status: DONE - repair required (see restore docs) Start time: 16 Feb 23 17:17:29 CET End time: 16 Feb 23 17:18:44 CET Duration: 1m15s Progress: 100% | 100% Snapshot Tag: sm_20230216160631UTC ╭───────────────┬─────────────┬────────┬─────────┬────────────┬──────────────┬────────╮ │ Keyspace │ Progress │ Size │ Success │ Downloaded │ Deduplicated │ Failed │ ├───────────────┼─────────────┼────────┼─────────┼────────────┼──────────────┼────────┤ │ test_keyspace │ 100% | 100% │ 7.938M │ 7.938M │ 7.938M │ 0 │ 0 │ ╰───────────────┴─────────────┴────────┴─────────┴────────────┴──────────────┴────────╯ Keyspace: test_keyspace ╭────────┬─────────────┬────────┬─────────┬────────────┬──────────────┬────────┬────────────────────────┬────────────────────────╮ │ Table │ Progress │ Size │ Success │ Downloaded │ Deduplicated │ Failed │ Started at │ Completed at │ ├────────┼─────────────┼────────┼─────────┼────────────┼──────────────┼────────┼────────────────────────┼────────────────────────┤ │ table1 │ 100% | 100% │ 2.642M │ 2.642M │ 2.642M │ 0 │ 0 │ 16 Feb 23 17:17:35 CET │ 16 Feb 23 17:18:37 CET │ ├────────┼─────────────┼────────┼─────────┼────────────┼──────────────┼────────┼────────────────────────┼────────────────────────┤ │ table2 │ 100% | 100% │ 2.657M │ 2.657M │ 2.657M │ 0 │ 0 │ 16 Feb 23 17:17:31 CET │ 16 Feb 23 17:18:33 CET │ ├────────┼─────────────┼────────┼─────────┼────────────┼──────────────┼────────┼────────────────────────┼────────────────────────┤ │ table3 │ 100% | 100% │ 2.640M │ 2.640M │ 2.640M │ 0 │ 0 │ 16 Feb 23 17:17:39 CET │ 16 Feb 23 17:18:41 CET │ ╰────────┴─────────────┴────────┴─────────┴────────────┴──────────────┴────────┴────────────────────────┴────────────────────────╯
Perform a dry run of a restore¶
We recommend performing a dry run of a restore prior to scheduling the restore.
It’s a useful way to verify whether all necessary prerequisites are fulfilled.
To perform a dry run, add the --dry-run parameter at the end of the restore command. If it works, you can remove the parameter from the command and schedule the restore without making any other changes.
A dry run verifies if nodes can access the provided backup location. If the location is not accessible, an error message will be displayed, and the restore is not scheduled.
sctool restore -c prod-cluster -L s3:test-bucket -T snapshot_tag --restore-tables --dry-run
Error: failed to get backup target: location is not accessible
192.168.100.23: failed to access s3:test-bucket make sure that the location is correct and credentials are set
192.168.100.22: failed to access s3:test-bucket make sure that the location is correct and credentials are set
192.168.100.21: failed to access s3:test-bucket make sure that the location is correct and credentials are set
Dry run also verifies if any backup contents are matched by given glob patterns. If not, an error message will be displayed, and restore is not scheduled.
sctool restore -c prod-cluster -L s3:test-bucket -T snapshot_tag -K non_existing_keyspace --restore-tables --dry-run
Error: get restore units: no data in backup locations match given keyspace pattern
Performing a dry run allows you to resolve all configuration or access issues before executing an actual restore.
If the dry run completes successfully, a summary of the restore is displayed. For example:
sctool restore -c prod-cluster -L s3:test-backup -T snapshot_tag -K test_keyspace --restore-tables --dry-run
NOTICE: dry run mode, restore is not scheduled
Keyspaces:
- test_keyspace: 7.938M (table2: 2.657M, table1: 2.642M, table3: 2.640M)
Disk size: ~7.938M
Locations:
- s3:test-backup
Snapshot Tag: sm_20230216160631UTC
Batch Size: 2
Parallel: 1