Skip to content

[opt](routine-load) enhance auto resume to keep routine load stable #32239

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
liaoxin01 merged 1 commit into from
Mar 18, 2024
Merged

[opt](routine-load) enhance auto resume to keep routine load stable #32239

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions fe/fe-core/src/main/java/org/apache/doris/common/InternalErrorCode.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,6 +35,7 @@ public enum InternalErrorCode {
TOO_MANY_FAILURE_ROWS_ERR(102),
CREATE_TASKS_ERR(103),
TASKS_ABORT_ERR(104),
CANNOT_RESUME_ERR(105),

// for MoW table
DELETE_BITMAP_LOCK_ERR(301);
Expand Down
6 changes: 3 additions & 3 deletions fe/fe-core/src/main/java/org/apache/doris/load/routineload/RoutineLoadJob.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1169,7 +1169,7 @@ public void afterAborted(TransactionState txnState, boolean txnOperated, String
+ " with reason: " + txnStatusChangeReasonString
+ " please check the jsonpaths";
updateState(JobState.PAUSED,
new ErrorReason(InternalErrorCode.TASKS_ABORT_ERR, msg),
new ErrorReason(InternalErrorCode.CANNOT_RESUME_ERR, msg),
false /* not replay */);
return;
case OFFSET_OUT_OF_RANGE:
Expand All @@ -1182,14 +1182,14 @@ public void afterAborted(TransactionState txnState, boolean txnOperated, String
+ " using the Alter ROUTINE LOAD command to modify it,"
+ " and resume the job";
updateState(JobState.PAUSED,
new ErrorReason(InternalErrorCode.TASKS_ABORT_ERR, msg),
new ErrorReason(InternalErrorCode.CANNOT_RESUME_ERR, msg),
false /* not replay */);
return;
case PAUSE:
msg = "be " + taskBeId + " abort task "
+ "with reason: " + txnStatusChangeReasonString;
updateState(JobState.PAUSED,
new ErrorReason(InternalErrorCode.TASKS_ABORT_ERR, msg),
new ErrorReason(InternalErrorCode.CANNOT_RESUME_ERR, msg),
false /* not replay */);
return;
default:
Expand Down
5 changes: 4 additions & 1 deletion fe/fe-core/src/main/java/org/apache/doris/load/routineload/ScheduleRule.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -63,7 +63,10 @@ public static boolean isNeedAutoSchedule(RoutineLoadJob jobRoutine) {
jobRoutine.id, jobRoutine.firstResumeTimestamp, jobRoutine.autoResumeCount,
jobRoutine.pauseReason == null ? "null" : jobRoutine.pauseReason.getCode().name());
}
if (jobRoutine.pauseReason != null && jobRoutine.pauseReason.getCode() == InternalErrorCode.REPLICA_FEW_ERR) {
if (jobRoutine.pauseReason != null
&& jobRoutine.pauseReason.getCode() != InternalErrorCode.MANUAL_PAUSE_ERR
&& jobRoutine.pauseReason.getCode() != InternalErrorCode.TOO_MANY_FAILURE_ROWS_ERR
&& jobRoutine.pauseReason.getCode() != InternalErrorCode.TOO_MANY_FAILURE_ROWS_ERR) {
int dead = deadBeCount();
if (dead > Config.max_tolerable_backend_down_num) {
if (LOG.isDebugEnabled()) {
Expand Down
2 changes: 1 addition & 1 deletion fe/fe-core/src/main/java/org/apache/doris/service/FrontendServiceImpl.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1981,7 +1981,7 @@ public TStreamLoadMultiTablePutResult streamLoadMultiTablePut(TStreamLoadPutRequ
try {
RoutineLoadJob routineLoadJob = Env.getCurrentEnv().getRoutineLoadManager()
.getRoutineLoadJobByMultiLoadTaskTxnId(request.getTxnId());
routineLoadJob.updateState(JobState.PAUSED, new ErrorReason(InternalErrorCode.MANUAL_PAUSE_ERR,
routineLoadJob.updateState(JobState.PAUSED, new ErrorReason(InternalErrorCode.CANNOT_RESUME_ERR,
"failed to get stream load plan, " + exception.getMessage()), false);
} catch (UserException e) {
LOG.warn("catch update routine load job error.", e);
Expand Down