Details
-
New Feature
-
Status: Open (View Workflow)
-
Major
-
Resolution: Unresolved
-
None
-
None
-
None
Description
Given the logline:
Jul 7 15:00:07 mcs1-20250618 joblist[6117]: 07.109616 |163|0|0| C 05 CAL0000: TupleAggregateStep::doThreadedAggregate() MCS-2056: There was an IO error during a disk-based aggregation: No such file or directory %%10%% |
|
The usual root causes not match.
It happens with 23.10.4
NO UDAFs are involved in the query.
No GROUP_CONCAT or JSON_ARRAYAGG is used
There is only one place , where this error can be triggered
The catch block
catch (...) |
{
|
handleException(std::current_exception(), logging::tupleAggregateStepErr,
|
logging::ERR_AGGREGATION_TOO_BIG, "TupleAggregateStep::doThreadedAggregate()"); |
fEndOfResult = true; |
}
|
|
from
uint64_t TupleAggregateStep::doThreadedAggregate(ByteStream& bs, RowGroupDL* dlp)
in
tupleaggregatestep.cpp
We might have here a new possibility, why this error occur,
where the message "No such file or directory" is not sufficient to explain the error.
Also
ERR_AGGREGATION_TOO_BIG
must be correct, so more information must be logged to investigate the root cause here.