From f09a98b7f5c84d834f4073da1bc751e231c0c13c Mon Sep 17 00:00:00 2001 From: xrendan Date: Thu, 10 Sep 2026 11:45:26 -0600 Subject: [PATCH 1/2] Point the worker's agent at the web service inside the compose network AgentEvaluateCommitmentJob and AgentProcessEntryJob shell out to the claude CLI, which writes results back via RAILS_API_URL. The worker container never had that variable, so the subprocess defaulted to http://localhost:3000 and could not reach the web service. --- docker-compose.prod.yml | 3 +++ 1 file changed, 3 insertions(+) diff --git a/docker-compose.prod.yml b/docker-compose.prod.yml index f3246bb..7934a2d 100644 --- a/docker-compose.prod.yml +++ b/docker-compose.prod.yml @@ -40,6 +40,9 @@ services: - HOST - RAILS_LOG_LEVEL - GOOD_JOB_EXECUTION_MODE=external + # The agent subprocess posts results back to the Rails API; inside the compose + # network the web service is reachable by name, not on localhost. + - RAILS_API_URL=http://web:3000 - SMTP_USER_NAME - SMTP_PASSWORD - SMTP_ADDRESS From 083b311151b2bbf2637825264c2c60594a649430 Mon Sep 17 00:00:00 2001 From: xrendan Date: Thu, 10 Sep 2026 11:45:50 -0600 Subject: [PATCH 2/2] Raise the agent job enqueue limit above the commitment count GoodJob drops enqueues beyond enqueue_limit without raising. With 598 non-broken commitments and a limit of 550, AgentWeeklyScanJob silently skipped the tail of every full scan. --- app/jobs/agent_evaluate_commitment_job.rb | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/app/jobs/agent_evaluate_commitment_job.rb b/app/jobs/agent_evaluate_commitment_job.rb index 5d9c9c0..215bf7a 100644 --- a/app/jobs/agent_evaluate_commitment_job.rb +++ b/app/jobs/agent_evaluate_commitment_job.rb @@ -6,7 +6,7 @@ class AgentEvaluateCommitmentJob < ApplicationJob include GoodJob::ActiveJobExtensions::Concurrency good_job_control_concurrency_with( perform_limit: 5, - enqueue_limit: 550, + enqueue_limit: 2000, key: "AgentEvaluateCommitmentJob" )