Skip to content

Set assignment state to "started" in case of zero allocations #115824

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
merged 1 commit into from
Oct 29, 2024
Merged

Set assignment state to "started" in case of zero allocations #115824

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
3 changes: 3 additions & 0 deletions ...ain/java/org/elasticsearch/xpack/core/ml/inference/assignment/TrainedModelAssignment.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -533,6 +533,9 @@ public AssignmentState calculateAssignmentState() {
if (assignmentState.equals(AssignmentState.STOPPING)) {
return assignmentState;
}
if (taskParams.getNumberOfAllocations() == 0) {
return AssignmentState.STARTED;
Copy link
Contributor Author

@jan-elastic jan-elastic Oct 29, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

There are code paths where you can get into state of numberOfAllocations=0 and assignmentState=STARTING. I think that can happen when the scheduler moves jobs around, which triggers the state to be recalculated here.

This causes issues, because assignments in a STARTING state cannot be updated, and an assignment with numberOfAllocations=0 won't get out of that state.

}
if (nodeRoutingTable.values().stream().anyMatch(r -> r.getState().equals(RoutingState.STARTED))) {
return AssignmentState.STARTED;
}
Expand Down
5 changes: 5 additions & 0 deletions ...ava/org/elasticsearch/xpack/core/ml/inference/assignment/TrainedModelAssignmentTests.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -172,6 +172,11 @@ public void testCalculateAssignmentState_GivenNoStartedAssignments() {
assertThat(builder.calculateAssignmentState(), equalTo(AssignmentState.STARTING));
}

public void testCalculateAssignmentState_GivenNumAllocationsIsZero() {
TrainedModelAssignment.Builder builder = TrainedModelAssignment.Builder.empty(randomTaskParams(0), null);
assertThat(builder.calculateAssignmentState(), equalTo(AssignmentState.STARTED));
}

public void testCalculateAssignmentState_GivenOneStartedAssignment() {
TrainedModelAssignment.Builder builder = TrainedModelAssignment.Builder.empty(randomTaskParams(5), null);
builder.addRoutingEntry("node-1", new RoutingInfo(4, 4, RoutingState.STARTING, ""));
Expand Down
Loading