Skip to content

feat: enable StuckDetector for running task auto-recovery #1067

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

Open
sh1nj1 wants to merge 2 commits into
base: main
Choose a base branch
from
Open

feat: enable StuckDetector for running task auto-recovery #1067

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
4fe7ab0
feat: enable StuckDetector and add pending_approval timeout
sh1nj1 Mar 27, 2026
86e4513
refactor: remove pending_approval auto-cancel from StuckDetector
sh1nj1 Apr 15, 2026
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
2 changes: 1 addition & 1 deletion engines/collavre/app/services/collavre/orchestration/policy_resolver.rb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,7 +38,7 @@ class PolicyResolver
"token_spike_window_minutes" => 10
},
"stuck_detection" => {
"enabled" => false,
"enabled" => true,
"task_stuck_threshold_minutes" => 30, # Task running for > N minutes
"creative_stall_threshold_minutes" => 120, # Creative no progress for > N minutes
"create_system_comment" => true # Create system comment on escalation
Expand Down
5 changes: 2 additions & 3 deletions engines/collavre/app/services/collavre/orchestration/stuck_detector.rb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -62,9 +62,10 @@ def auto_recover_stuck_tasks(stuck_items)
next unless task.status == "running"

task.update!(status: "failed")
minutes_stuck = ((Time.current - stuck_item.stuck_since) / 60).round
Rails.logger.info(
"[StuckDetector] Auto-recovered task #{task.id} (agent=#{task.agent_id}): " \
"marked as failed after #{((Time.current - stuck_item.stuck_since) / 60).round} minutes"
"marked as failed after #{minutes_stuck} minutes"
)

# Release resources held by the stuck task
Expand Down Expand Up @@ -93,7 +94,6 @@ def detect_stuck_tasks(config)
.where("updated_at < ?", threshold_time)

stuck_tasks.filter_map do |task|
# Skip if already escalated recently
next if recently_escalated?(task)

escalation_targets = find_escalation_targets(task)
Expand Down Expand Up @@ -254,7 +254,6 @@ def build_message(stuck_item)
when :task
task = stuck_item.item
minutes_stuck = ((Time.current - stuck_item.stuck_since) / 60).round

[
"collavre.stuck_detection.task_stuck",
{
Expand Down