Manage migrations
Manage your migrations with LiveData Migrator using either the UI or CLI.
caution
Do not write to target filesystem paths when a migration is underway. This could interfere with LiveData Migrator functionality and lead to undetermined behavior.
Use different filesystem paths when writing to the target storage directly (and not through LiveData Migrator).
#
Manage migrations with the UIThe Dashboard displays an overview of migrations and their status, showing what pre-existing data has been moved and data added since the migration started.
Click View migration to see more detail.
note
A migration must be stopped or aborted before it can be deleted.
#
Migration metricsView details about migration events and remaining scanned paths to better understand a migration's progress.
- In the dashboard, select a migration.
- You will be taken to the Status of the migration tab by default.
Under Migration Metrics, observe the following:
- Events yet to be migrated: this graph shows the size of the scheduled actions queue across a time period selected in the Filter by dropdown menu.
- Paths to be scanned: this graph displays the size of the queue for paths yet to be scanned across a time period selected in the Filter by dropdown menu.
#
Assign exclusions to an existing migrationAdding exclusions to an existing migration will change the future actions performed for that migration, but will not affect previously migrated content.
- In the dashboard, select a migration.
- Under the sidebar that appears, select Exclusions.
- Click Add and select the appropriate exclusion template from the drop-down list.
#
Remove exclusions from an existing migration- In the dashboard, select a migration.
- Under the sidebar that appears, select Exclusions.
- Remove any of the exclusions one at a time.
#
Start a migrationTo start a migration and initiate data transfer, click the Start button.
#
Stop a migrationWhen you start a migration, the Start button will be replaced with a Stop button. Click it to stop the migration and suspend all current file transfer.
#
Resume a migrationWhen you stop a migration, the Stop button will be replaced with a Resume button. Click it to resume the migration, which will continue from where it left off.
#
Delete a migrationnote
Migrations must be stopped before they can be deleted.
When viewing a migration, click on the Details panel. Click the option to Delete the migration.
#
Manage migrations with the CLICommand | Action |
---|---|
migration stop | Stop a migration |
migration resume | Resume a stopped migration |
migration delete | Delete a migration |
migration exclusion add | Add an exclusion to a migration |
migration exclusion delete | Remove an exclusion from a migration |
migration list | List running and active migrations |
migration start | Start a migration |
migration show | Get migration details |
migration pending-region add | Add a pending region to a migration |
status | Get migration status |
#
Pending regionsLiveData Migrator uses pending regions to keep your directories up to date if they change during data migrations. If directories are updated while being migrated, the changed paths on the source filesystem are tracked by the migration so they can be re-scanned for updates when the migration finishes.
LiveData Migrator collects pending regions automatically during a migration, but you can manually add them if you want the directories to be re-scanned after further updates. You can also re-run the entire migration by making the root directory the pending region.
#
Add pending regions in the UITo add a pending region in the UI:
- Click on the migration you want to add a pending region to in the dashboard.
- Under the sidebar that appears, select Rescan.
- Type the path to the directory in Add rescan directories.
- Choose Overwrite or Skip If Size Match.
- Click +Add.
#
Add pending regions in the CLIAdd a pending region to a migration by running the migration pending-region add
command.
migration pending-region add --migration-id myFirstMigration --path /dir1/userA
#
Data migration statesMigrations can be in one of eight states:
NONSCHEDULED
: A non-scheduled migration has been defined but not yet started. Create a migration in this state by not specifying the --auto-start
parameter on creation.
SCHEDULED
: A scheduled migration will start when directed to run.
STARTING
: A starting migration is being started and will soon begin transferring content to the target.
RUNNING
: A running migration is scanning through source content and transferring content to the target, as well as responding to change notifications from the source if applicable.
LIVE
: A live migration has completed scanning through source content, continues to respond to change notifications from the source, and will transfer content to and make changes in the target as required.
PAUSING
: A pausing migration has been instructed to pause transfer, but is temporarily continuing to make changes to the target.
PAUSED
: A paused migration has been instructed to pause transfer, and is not transferring content or making other changes to the target.
ABORTED
: An aborted migration will not make any changes to the target and cannot be run again.
COMPLETED
: A completed migration has scanned through all source content and finished transferring all applicable data to the target filesystem, and has ignored any updates to the source data.
#
Monitoring Failed OperationsThe Failed Operations panel on LiveData UI's Migration screen provides a counter of any failed operations, including paths that failed to migrate. For a breakdown of the latest failures, click View details. This opens a list of failures that includes the following information for each failure:
Date and time: Date and time of the failure.
Example2020-01-07 13:30:40 AM UTC
Path: Path that failed to migrate.
Example/path/to/migration/files/000000_1
Failure: Error message generated by the failure.
Examplejava.io.IOException: Error rescanning [/path/to/migration/files/part-1234] [not found]
note
Historical failure information may be garbage collected, so it may not be possible to show all failures. In this case a warning will appear that states' "We're unable to show detailed results of failed operations. We only have detailed results from the most recent [number of available failures] results"