Out of curiosity, is running a query against a system table (such as
dual) supposed to exclusively acquire resources from the
sysquery pool? It appears to be also dipping into the general pool:
dbadmin=> select * from query_requests where session_id = 'v7101n2-1244:0xaa6' and statement_id = 2;
-[ RECORD 1 ]-------+-------------------------------------------------- node_name | v_vmart_node0002 user_name | dbadmin session_id | v7101n2-1244:0xaa6 request_id | 2 transaction_id | 49539595901136378 statement_id | 2 request_type | QUERY request | select 1 from dual; request_label | search_path | "$user", public, v_catalog, v_monitor, v_internal memory_acquired_mb | 100 success | t error_count | start_timestamp | 2014-12-22 13:22:08.802066-06 end_timestamp | 2014-12-22 13:22:08.813208-06 request_duration_ms | 11 is_executing | f dbadmin=> select * from dc_resource_acquisitions where transaction_id = '49539595901136378' and statement_id = 2; -[ RECORD 1 ]----+------------------------------ time | 2014-12-22 13:22:08.803435-06 node_name | v_vmart_node0002 transaction_id | 49539595901136378 statement_id | 2 request_type | Acquire pool | 45035996273720462 pool_name | general priority | 0 memory_kb | 102400 filehandles | 0 threads | 0 queries | 0 start_time | 2014-12-22 13:22:08.802436-06 succeeded | t result | Granted failing_resource | is_required | t -[ RECORD 2 ]----+------------------------------ time | 2014-12-22 13:22:08.811576-06 node_name | v_vmart_node0002 transaction_id | 49539595901136378 statement_id | 2 request_type | Reserve pool | 45035996273720464 pool_name | sysquery priority | 110 memory_kb | 4231 filehandles | 0 threads | 4 queries | 1 start_time | 2014-12-22 13:22:08.811543-06 succeeded | t result | Granted failing_resource | is_required | t
