API Doc
Search Docs...
⌘ K
OverviewDeploymentManagementOperationReferenceGlossary
    ACOS 6.3.0
  • Arcfra Cloud Operating System>
  • CLI command>
  • Managing cluster internal I/Os>
  • Managing data migration and recovery

Managing data migration

Scanning for extents that require migration

Procedure

Run the following command on any node in the cluster:

zbs-meta migrate scan_immediate

Output note

If the command succeeds, no output is displayed.

Enabling data migration

Procedure

Run the following command on any node in the cluster:

zbs-meta migrate enable

Output example

Enabled

Disabling data migration

Procedure

Run the following command on any node in the cluster:

zbs-meta migrate disable

Output example

Disabled

Querying the current data migration mode

Procedure

Run the following command on any node in the cluster:

zbs-meta migrate get_mode_info

Output example

Migrate enable: True
Migrate scan interval: 120s
Perf thick ratio: 0.722, load: LOAD_HIGH
Perf thin  ratio: 0.494, load: LOAD_LOW
Cap        ratio: 0.346, load: LOAD_LOW

Output note

ParameterDescription
Migrate enableWhether data migration is enabled.
Migrate scan intervalThe scan interval for data migration.
Perf thick loadThe load status of the pinned space.
Perf thin loadThe load status of the performance tier space.
Cap loadThe load status of the capacity tier space.

Viewing extents currently under migration

Procedure

Run the following command on any node in the cluster:

zbs-meta migrate list

Output example

No migration command is running.

Migrating a specified extent

Procedure

Run the following command on any node in the cluster:

zbs-meta migrate pextent <pid> <src_cid> <dst_cid> <replace_cid> [--ignore_topo_safety]

Parameter Description
src_cid The chunk ID of the source.
dst_cid The chunk ID of the destination.
replace_cid The chunk ID to be replaced.
--ignore_topo_safety Whether to ignore topology security for this migration, defaults to False.

Output note

If the command succeeds, no output is displayed.

Locally aggregating a specified virtual volume

Procedure

Execute the following command on any cluster node to manually migrate extents of the specified virtual volume to their preferred local nodes, respectively:

zbs-meta migrate volume <volume_id> [--ignore_perf] [--ignore_cap]

Parameter Description
volume_id The ID of the specified virtual volume.
--ignore_perf Ignores extents in the performance tier.
--ignore_cap Ignores extents in the capacity tier.

Usage example

$ zbs-meta migrate volume 135a8de9-64f7-4e89-a81c-844976d871be --ignore_perf

Output example

need_migrate_num: 10

This output indicates that 10 extents on the virtual volume are migrated to their preferred local nodes, respectively.

Clearing manually triggered migration commands

Procedure

Run the following command on any node in the cluster:

zbs-meta migrate clear_manual_migrate

The cluster will not automatically generate the migration command before a manually triggered migration command is generated. In this case, you can use this command to restore automatic system migration.

Output example

If the command succeeds, no output is displayed.

Querying the reporting configuration of extents to be migrated on unmounted disks

Procedure

Run the following command on any node in the cluster:

zbs-meta migrate show_umounting_report_option

Output example

Report enable:  True
Include cache:  False
Include partition:  True
Max report size:  2048

Output note

ParameterDescription
Report enableWhether chunks in the cluster report extents to be migrated for meta-assisted migration.
Include cacheWhether chunks in the cluster report extents to be migrated from the performance tier. The default value is False.
Include partitionWhether chunks in the cluster report extents to be migrated from the capacity tier. The default value is True.
Max report sizeThe maximum number of migratable extents that can be reported in a single report by chunks in the cluster.

Updating the reporting configuration of extents to be migrated on unmounted disks

Procedure

Run the following command on any node in the cluster:

zbs-meta migrate update_umounting_report_option [--report_enable] [--include_cache] [--include_partition] [--max_report_size]

Output note

If the command succeeds, no output is displayed.

Usage example

$ zbs-meta migrate update_umounting_report_option --include_partition True --max_report_size 1024

Querying extents to be migrated on unmounted disks

Procedure

Run the following command on any cluster node to view the extents to be migrated on unmounted disks reported by chunks:

zbs-meta migrate list_umounting_pextents [--cid CID] [--length LENGTH] [--ignore_ec] [--show_details]

Parameter Description
--cid The chunk ID. If not specified, migratable extents on unmounted disks of all chunks are queried.
--length The number of PExtents to return. Default and maximum: 1024.
--ignore_ec Whether to ignore extents with redundancy type EC. The default value is False.
--show_details Whether to display detailed information about the extents to be migrated on unmounted disks. By default, details are not displayed.

Usage example

$ zbs-meta migrate list_umounting_pextents --cid 4 --show_details

Output example

Umounting cid: 4, cap report size: 5
    Pid    Umounting Cid  Pextent Type    Resiliency Type    Thin Provision
-------  ---------------  --------------  -----------------  ----------------
2798982                4  PT_CAP          RT_REPLICA         True
2822009                4  PT_CAP          RT_EC              True
2822033                4  PT_CAP          RT_EC              True
2937914                4  PT_CAP          RT_REPLICA         True
2954389                4  PT_CAP          RT_EC              True

Adjusting migration speed

Refer to Setting internal I/O modes and priorities.