Maintenance Command
FLUSH
Persist all the data points in the memory table of the storage group to the disk, and seal the data file.
Note: This command does not need to be invoked manually by the client. IoTDB has WAL to ensure data security and IoTDB will flush when appropriate. Frequently call flush can result in small data files that degrade query performance.
IoTDB> FLUSH
IoTDB> FLUSH root.ln
IoTDB> FLUSH root.sg1,root.sg2
MERGE
Execute Level Compaction and unsequence Compaction task. Currently IoTDB supports the following two types of SQL to manually trigger the compaction process of data files:
MERGE
Execute the level compaction first and then execute the unsequence compaction. In unsequence compaction process, this command is executed very fast by rewriting the overlapped Chunks only, while there is some redundant data on the disk eventually.FULL MERGE
Execute the level compaction first and then execute the unsequence compaction. In unsequence compaction process, this command is executed slow due to it takes more time to rewrite all data in overlapped files. However, there won’t be any redundant data on the disk eventually.
IoTDB> MERGE
IoTDB> FULL MERGE
CLEAR CACHE
Clear the cache of chunk, chunk metadata and timeseries metadata to release the memory footprint.
IoTDB> CLEAR CACHE
SET STSTEM TO READONLY / WRITABLE
Manually set IoTDB system to read-only or writable mode.
IoTDB> SET SYSTEM TO READONLY
IoTDB> SET SYSTEM TO WRITABLE
SCHEMA SNAPSHOT
To speed up restarting of IoTDB, users can create snapshot of schema and avoid recovering schema from mlog file. This feature doesn’t support scenarios involving Schema Template, Tag/Attribute, or Aligned Timeseries.
IoTDB> CREATE SNAPSHOT FOR SCHEMA
Timeout
IoTDB supports session and query level timeout.
Session timeout
Session timeout controls when idle sessions are closed. An idle session is one that had not initiated any query or non-query operations for a period of time.
Session timeout is disabled by default and can be set using the session_timeout_threshold
parameter in IoTDB configuration file.
Query timeout
For queries that take too long to execute, IoTDB will forcibly interrupt the query and throw a timeout exception, as shown in the figure:
IoTDB> select * from root;
Msg: 701 Current query is time out, please check your statement or modify timeout parameter.
The default timeout of a query is 60000 ms,which can be customized in the configuration file through the query_timeout_threshold
parameter.
If you use JDBC or Session, we also support setting a timeout for a single query(Unit: ms):
((IoTDBStatement) statement).executeQuery(String sql, long timeoutInMS)
session.executeQueryStatement(String sql, long timeout)
If the timeout parameter is not configured or with a negative number, the default timeout time will be used. If value 0 is used, timeout function will be disabled.
Query abort
In addition to waiting for the query to time out passively, IoTDB also supports stopping the query actively:
KILL QUERY <queryId>
You can abort the specified query by specifying queryId
. If queryId
is not specified, all executing queries will be killed.
To get the executing queryId
,you can use the show query processlist
command,which will show the list of all executing queries,with the following result set:
Time | queryId | statement |
---|---|---|
The maximum display length of statement is 64 characters. For statements with more than 64 characters, the intercepted part will be displayed.