Loading...
Searching...
No Matches
NYT::TUserOperationSpecBase< TDerived > Struct Template Reference
Base spec for all operations with user jobs. More...
#include <operation.h>
Inheritance diagram for NYT::TUserOperationSpecBase< TDerived >:
Public Member Functions | |
| TSelf & | FailOnJobRestart (bool) |
| On any unsuccessful job completion (i.e. abortion or failure) force the whole operation to fail. | |
| TSelf & | StderrTablePath (TYPath) |
| Table to save whole stderr of operation. | |
| TSelf & | CoreTablePath (TYPath) |
| Table to save coredumps of operation. | |
| TSelf & | WaitingJobTimeout (TDuration) |
| How long should the scheduler wait for the job to be started on a node. | |
| Public Member Functions inherited from NYT::TOperationSpecBase< TDerived > | |
| TSelf & | TimeLimit (TDuration) |
| Limit on operation execution time. | |
| TSelf & | Alias (TString) |
| Alias for searching for an operation in the future. | |
| TSelf & | Title (TString) |
| Title to be shown in web interface. | |
| TSelf & | Pool (TString) |
| Pool to be used for this operation. | |
| TSelf & | Weight (double) |
| Weight of operation. | |
| FLUENT_OPTIONAL_VECTOR_FIELD_ENCAPSULATED (TString, PoolTree) | |
| Pool tree list that operation will use. | |
| FLUENT_FIELD_OPTION_ENCAPSULATED (TSchedulerResources, ResourceLimits) | |
| How much resources can be consumed by operation. | |
| TSelf & | MaxFailedJobCount (ui64) |
| How many jobs can fail before operation is failed. | |
| TSelf & | Annotations (TNode) |
| TSelf & | Description (TNode) |
Detailed Description
template<class TDerived>
struct NYT::TUserOperationSpecBase< TDerived >
struct NYT::TUserOperationSpecBase< TDerived >
Base spec for all operations with user jobs.
Member Function Documentation
◆ CoreTablePath()
template<class TDerived>
| TSelf & NYT::TUserOperationSpecBase< TDerived >::CoreTablePath | ( | TYPath | ) |
Table to save coredumps of operation.
◆ StderrTablePath()
template<class TDerived>
| TSelf & NYT::TUserOperationSpecBase< TDerived >::StderrTablePath | ( | TYPath | ) |
Table to save whole stderr of operation.
◆ WaitingJobTimeout()
template<class TDerived>
| TSelf & NYT::TUserOperationSpecBase< TDerived >::WaitingJobTimeout | ( | TDuration | ) |
How long should the scheduler wait for the job to be started on a node.
When you run huge jobs that require preemption of all the other jobs on a node, the default timeout might be insufficient and your job may be aborted with 'waiting_timeout' reason. This is especially problematic when you are setting 'FailOnJobRestart' option.
- Note
- The value must be between 10 seconds and 10 minutes.
The documentation for this struct was generated from the following file:
- yt/cpp/mapreduce/interface/operation.h
