LOAD_SOURCES
Like LOAD_STREAMS, monitors active and historical load metrics on each node. The LOAD_SOURCES table breaks information down by source and portion. Rows appear in this table only for COPY operations that are profiled or run for more than one second. LOAD_SOURCES does not record information about loads from ORC or Parquet files or COPY LOCAL.
A row is added to this table when the loading of a source or portion begins. Column values related to the progress of the load are updated during the load operation.
Columns that uniquely identify the load source (the various ID and name columns) and column IS_EXECUTING always have non-NULL values.
Column Name | Data Type | Description |
---|---|---|
SESSION_ID | VARCHAR | Identifier of the session for which Vertica captures load stream information. This identifier is unique within the cluster for the current session but can be reused in a subsequent session. |
TRANSACTION_ID | INTEGER | Identifier for the transaction within a session. If a session is active, but no transaction has begun, this value is NULL. |
STATEMENT_ID | INTEGER | Unique numeric ID for the currently running statement. NULL indicates that no statement is currently being processed. The combination of TRANSACTION_ID, STATEMENT_ID uniquely identifies a statement within a session. |
STREAM_NAME | VARCHAR |
Load stream identifier. If the user does not supply a specific name, the STREAM_NAME default value is
This system table includes stream names for every COPY statement that takes more than 1 second to run. The 1-second duration includes the time to plan and execute the statement. |
SCHEMA_NAME | VARCHAR | Schema name for which load information is listed. Lets you identify two streams that are targeted at tables with the same name in different schemas. NULL, if selecting from an external table. |
TABLE_OID | INTEGER | A unique numeric ID assigned by the Vertica catalog that identifies the table. NULL, if selecting from an external table. |
TABLE_NAME | VARCHAR | Name of the table being loaded. NULL, if selecting from an external table. |
NODE_NAME | VARCHAR | Name of the node loading the source. |
SOURCE_NAME | VARCHAR |
|
PORTION_OFFSET | INTEGER | Offset of the source portion, or NULL if not apportioned. |
PORTION_SIZE | INTEGER | Size of the source portion, or NULL if not apportioned. |
IS_EXECUTING | BOOLEAN | Whether this source is currently being parsed, where t is true and f is false. |
READ_BYTES | INTEGER | Number of bytes read from the input file. |
ROWS_PRODUCED | INTEGER | Number of rows produced from parsing the source. |
ROWS_REJECTED | INTEGER | Number of rows rejected from parsing the source. If CopyFaultTolerantExpressions is true, also includes rows rejected during expression evaluation. |
INPUT_SIZE | INTEGER |
Size of the input source in bytes, or NULL for unsized sources. For UDSources, this value is the value returned by getSize() . |
PARSE_COMPLETE_PERCENT | INTEGER | Percent of rows from the input file that have been parsed. |
FAILURE_REASON | VARCHAR |
Indicates cause for failure, one of the following:
In all other cases, set to NULL. |
PEAK_COOPERATING_THREADS | INTEGER | The peak number of threads parsing this source in parallel. |
CLOCK_TIME_SOURCE | INTEGER | Displays in real-time how many microseconds (µs) have been consumed by the UDSource phase of a load operation. |
CLOCK_TIME_FILTERS | INTEGER | Displays in real-time how many microseconds (µs) have been consumed by all UDFilter phases of a load operation. |
CLOCK_TIME_CHUNKER | INTEGER | Displays in real-time how many microseconds (µs) have been consumed by the UDChunker phase of a load operation. |
CLOCK_TIME_PARSER | INTEGER | Displays in real-time how many microseconds (µs) have been consumed by the UDParser phase of a load operation. |
Privileges
Non-superuser: No explicit privileges required. You only see records for tables that you have privileges to view.