| #744 |
[FEATURE] Kafka Consumer Service Integration
Add native Kafka consumer service for event-driven workflows. Based on DBOS analysis - DBOS has @kaf...
|
closed |
high |
2025-12-30 02:37 |
- |
|
| #743 |
Centralize hardcoded timeout values in config.ini
Multiple timeout values are hardcoded across the codebase: claim_timeout (30s), timeout_threshold (6...
|
closed |
low |
2025-12-30 01:25 |
- |
|
| #742 |
Move TimeoutService from per-worker to system cron job
TimeoutService is currently started in every worker. Multiple workers scanning same queues for stuck...
|
closed |
high |
2025-12-30 01:25 |
- |
|
| #741 |
Move JanitorService from per-worker to system cron job
JanitorService is currently started in worker.py for highway_default queue workers. This means N wor...
|
closed |
high |
2025-12-30 01:25 |
- |
|
| #740 |
Verify Vault has LLM API keys for K8s deployment
After adding LLM API keys to Helm templates (api-deployment.yaml, worker-deployment.yaml), need to v...
|
closed |
medium |
2025-12-29 09:36 |
- |
|
| #739 |
API auto-extracts app metadata from source code
Fixed bad API design where UI had to parse Python to extract class_name and actions. API now auto-ex...
|
closed |
medium |
2025-12-29 06:26 |
- |
|
| #738 |
Shell security check incorrectly rejects commands in K8s pods
When workers run inside K8s pods (not using Docker-in-Docker sandbox), shell commands with && or || ...
|
closed |
high |
2025-12-29 04:23 |
- |
|
| #737 |
Enable network access in sandbox containers after Sysbox deployment
After Sysbox+DinD is deployed on srv2 workers, update sandbox to allow network access. Current: netw...
|
closed |
medium |
2025-12-29 01:24 |
- |
|
| #736 |
Event and IPC tests failing in production
Event/workflow tests fail with 'failed' status. Tests: test_standard_sleep_wake_event, test_final_ip...
|
closed |
high |
2025-12-29 01:00 |
- |
|
| #735 |
Datashard logging tests failing in production
3 datashard logging tests fail because workflows complete with 'failed' status. Tests: test_simple_w...
|
closed |
medium |
2025-12-29 01:00 |
- |
|
| #734 |
Circuit breaker returns UNKNOWN state in production
Circuit breaker tests fail because state returns 'UNKNOWN' instead of 'OPEN'. Tests: test_circuit_br...
|
closed |
low |
2025-12-29 01:00 |
- |
|
| #733 |
Artifact workflow tests stuck in pending state
2 artifact workflow tests stuck in 'pending' state after 30s timeout. Tests: test_workflow_with_arti...
|
closed |
high |
2025-12-29 01:00 |
- |
|
| #732 |
Docker tools tests failing in production K8s
All 9 docker tools integration tests fail with 'failed' status on production K8s. Workers likely don...
|
closed |
medium |
2025-12-29 01:00 |
- |
|
| #731 |
Production ansible playbook updates for multi-cluster deployment
## Task
Update ansible playbooks for production deployment across both clusters.
## Current State
-...
|
closed |
high |
2025-12-28 17:50 |
- |
|
| #730 |
Domain setup: highway.solutions + tilt.highway.rodmena.app
## Domains to configure
### Production: highway.solutions
- Points to: Production API + Dashboard
-...
|
closed |
high |
2025-12-28 17:50 |
- |
|
| #729 |
Vault/secrets handling for docker-compose dev environment
## Problem
Production uses Vault sidecar injection in K8s:
- vault.hashicorp.com/agent-inject annota...
|
closed |
high |
2025-12-28 17:50 |
- |
|
| #728 |
Tiltfile + docker-compose for local dev with hot reload
## Task
Create Tiltfile and docker-compose.dev.yml for local development with hot reload.
## Requir...
|
closed |
high |
2025-12-28 17:50 |
- |
|
| #727 |
DockerHub private repo setup for highway images
## Task
Setup private DockerHub repository for highway images.
## Details
- DockerHub username: rod...
|
closed |
high |
2025-12-28 17:49 |
- |
|
| #726 |
EPIC: Deployment Infrastructure Overhaul - Tilt + Docker Compose + Production Split
## Overview
Restructure deployment infrastructure to separate production and development environment...
|
closed |
high |
2025-12-28 17:49 |
- |
|
| #725 |
CRITICAL: Zombie transactions block task recovery when worker pods crash
## Summary
When a worker pod crashes during task execution, the PostgreSQL connection may stay open ...
|
closed |
critical |
2025-12-28 16:45 |
- |
|