getChunksMeta

getChunksMeta([chunkPath], [top=1024])

chunkPath (optional) is the DFS path to one or multiple database chunks. It supports wildcards %, * and ?.

top (optional) is a positive number indicating the maximum number of chunks in the output. The default value is 1024. If it is set to -1, the number of returned chunks is not limited.

Return metadata of specified database chunks on the local datanode. If chunkPath is not specified, return metadata of all database chunks on the local data node.

It returns a table containing the following columns:

• site: Alias of the node

• chunkId: The chunk ID

• path: The path to database chunks

• dfsPath: The DFS path to database chunks

• type: The partition type. 0 for file chunk; 1 for tablet chunk.

• flag: Flag for deletion. flag=1 means the chunk cannot be queried, but has not been removed from disk.

• size: The disk space (in bytes) occupied by the file chunk. It returns 0 for tablet chunks. Use the getTabletsMeta function to check the disk space tablet chunk occupies.

• version: the version number

• state: Chunk state.

  • 0 (final state): Transaction has been completed or rolled back.

  • 1 (before-commit state): The transaction is being executed on the chunk, such as writing or deleting data.

  • 2 (after-commit state): The transaction has been committed.

  • 3 (waiting-for-recovery state): When there is a version conflict or data corruption, this state occurs after the data node sends a recovery request to the controller and before the controller initiates the recovery.

  • 4 (in-recovery state): This state occurs after the controller receives the recovery request and initiates the recovery. Upon the completion of recovery, the chunk reverts to the final state (0).

• versionList: version list.

• resolved: Whether the transaction of the chunk is in the commit phase that needs to be resolved. True indicates the transaction is in the resolution phase or needs to be resolved; False indicates the transaction has already completed the resolution phase or it is not required.

if(existsDatabase("dfs://testDB")){
  dropDatabase("dfs://testDB")
}
db=database("dfs://testDB", VALUE, 1..10)

n=1000000
t=table(rand(1..10, n) as id, rand(100.0, n) as x)
db.createPartitionedTable(t, `p1, `id).append!(t)

n=2000000
t=table(rand(1..10, n) as id, rand(100.0, n) as x, rand(100, n) as y)
db.createPartitionedTable(t, `pt2, `id).append!(t)
getChunksMeta("/testDB%");