Region migration

Region migration allows users to move regions between the Datanode.

Warning

This feature is only available on GreptimeDB running on cluster mode and

• Using Kafka WAL
• Using shared storage (e.g., AWS S3)

Otherwise, you can't perform a region migration.

Figure out the MetaSrv leader

The region migration procedure is only allowed to submit to the MetaSrv leader; we need to figure out which MetaSrv is a leader.

curl {YOUR_META_SRV_SERVER}:3002/admin/leader
# 10.244.0.157:3002

Figure out the region distribution of the table.

To figure out a table's region distribution, we need to send a route query to the MetaSrv, using the YOUR_META_SRV_LEADER_SERVER address we obtained before.

curl {YOUR_META_SRV_LEADER_SERVER}:3002/admin/route?table_id={TABLE_ID}

# "{\"type\":\"physical\",\"region_routes\":[{\"region\":{\"id\":4668629450752,\"name\":\"\",\"partition\":{\"column_list\":[],\"value_list\":[\"\\\"MaxValue\\\"\"]},\"attrs\":{}},\"leader_peer\":{\"id\":2,\"addr\":\"greptimedb-datanode-2.greptimedb-datanode.my-greptimedb:4001\"},\"follower_peers\":[]}],\"version\":2}"

For example, we have the following region distribution:

region id	leader_peer	leader_peer_id
4668629450752	greptimedb-datanode-2.greptimedb-datanode.my-greptimedb:4001	2

Select a Datanode as the migration destination.

Warning

The region migration won't be performed if the from_peer_id equals the to_peer_id.

Remember, if you deploy the cluster via the GreptimeDB operator, the peer_id of Datanode always starts from 0. For example, if you have a 3 Datanode GreptimeDB cluster, the available peer_id will be 0,1,2.

Finally, you can do a Region migration request via the following HTTP request:

curl "{YOUR_META_SRV_LEADER_SERVER}:3002/admin/region-migration?region_id={REGION_ID}&from_peer_id={FROM_PEER_ID}&to_peer_id={TO_PEER_ID}&replay_timeout=5m"

Option	Description	Required
region_id	The region id.	Required
from_peer_id	The peer id of the migration source(Datanode).	Required
to_peer_id	The peer id of the migration destination(Datanode).	Required
replay_timeout	The timeout of replay data. e.g., 1h30m10s	Required