gohorsejobs/docs/WORKFLOWS.md

Workflows - GoHorse Jobs

Last Updated: 2026-02-16 Purpose: Deployment and operational workflows

---

Git Sync (GitHub -> Forgejo)

Push changes to both GitHub (origin) and Forgejo (pipe) remotes.

Flow

graph LR
    A[Local] -->|git push| B[GitHub origin]
    B -->|git pull| C[VPS]
    C -->|git push pipe| D[Forgejo pipe]

Commands

# 1. Push to GitHub (origin)
git push origin HEAD

# 2. Push to Forgejo (pipe)
git push pipe HEAD

Full Sync (from VPS)

# 1. Pull from GitHub
cd /root/gohorsejobs
git pull origin dev

# 2. Push to Forgejo
git push pipe dev

# 3. Sync other branches (optional)
git pull origin main && git push pipe main
git pull origin hml && git push pipe hml

---

Deploy Frontend

Deploy Frontend to Apolo server (dev.gohorsejobs.com).

Prerequisites

• Podman installed locally
• Access to Forgejo registry
• SSH access to Apolo server

Steps

# 1. Login to Forgejo Registry
podman login -u 'y@rede5.com.br' forgejo-gru.rede5.com.br

# 2. Build with Production URLs
cd /home/yamamoto/lab/gohorsejobs
podman build \
  --build-arg NEXT_PUBLIC_API_URL=https://api-tmp.gohorsejobs.com \
  --build-arg NEXT_PUBLIC_BACKOFFICE_URL=https://b-tmp.gohorsejobs.com \
  -t forgejo-gru.rede5.com.br/rede5/gohorsejobs-frontend:latest \
  -f frontend/Dockerfile frontend/

# 3. Push to Registry
podman push forgejo-gru.rede5.com.br/rede5/gohorsejobs-frontend:latest

# 4. Deploy on Apolo
ssh root@apolo 'podman pull forgejo-gru.rede5.com.br/rede5/gohorsejobs-frontend:latest && systemctl restart gohorsejobs-frontend-dev'

# 5. Verify
ssh root@apolo 'systemctl status gohorsejobs-frontend-dev --no-pager | tail -5'

Environment Reference

Variable	Production Value
NEXT_PUBLIC_API_URL	https://api-tmp.gohorsejobs.com
NEXT_PUBLIC_BACKOFFICE_URL	https://b-tmp.gohorsejobs.com
NEXT_PUBLIC_SEEDER_API_URL	https://seeder.gohorsejobs.com

---

Deploy Backend

Deploy Backend API to Apolo server (api-tmp.gohorsejobs.com).

Steps

# 1. Login to Forgejo Registry
podman login forgejo-gru.rede5.com.br

# 2. Build
cd backend
podman build -t forgejo-gru.rede5.com.br/rede5/gohorsejobs-backend:latest .

# 3. Push
podman push forgejo-gru.rede5.com.br/rede5/gohorsejobs-backend:latest

# 4. Deploy on Apolo
ssh root@apolo 'podman pull forgejo-gru.rede5.com.br/rede5/gohorsejobs-backend:latest && systemctl restart gohorsejobs-backend-dev'

---

Deploy Backoffice

Deploy Backoffice API to Apolo server (b-tmp.gohorsejobs.com).

Steps

# 1. Build
cd backoffice
podman build -t forgejo-gru.rede5.com.br/rede5/gohorsejobs-backoffice:latest .

# 2. Push
podman push forgejo-gru.rede5.com.br/rede5/gohorsejobs-backoffice:latest

# 3. Deploy
ssh root@apolo 'podman pull forgejo-gru.rede5.com.br/rede5/gohorsejobs-backoffice:latest && systemctl restart gohorsejobs-backoffice-dev'

---

Database Operations

Run Migrations

# Local
cd backend && go run ./cmd/manual_migrate

# Via SSH (remote)
ssh root@apolo 'cd /mnt/data/gohorsejobs/backend && ./migrate.sh'

Seed Database

# Trigger Seeder API
curl -X POST https://seeder.gohorsejobs.com/seed

# Local
cd seeder-api && npm run seed

Database Shell

# Via SSH
ssh root@apolo 'podman exec -it postgres-main psql -U yuki -d gohorsejobs_dev'

---

Service Management (Apolo)

View Status

ssh root@apolo 'systemctl status gohorsejobs-backend-dev --no-pager'
ssh root@apolo 'systemctl status gohorsejobs-frontend-dev --no-pager'
ssh root@apolo 'systemctl status gohorsejobs-backoffice-dev --no-pager'
ssh root@apolo 'systemctl status gohorsejobs-seeder-dev --no-pager'

Restart Services

# All services
ssh root@apolo 'systemctl restart gohorsejobs-backend-dev gohorsejobs-frontend-dev gohorsejobs-backoffice-dev gohorsejobs-seeder-dev'

# Individual service
ssh root@apolo 'systemctl restart gohorsejobs-backend-dev'

View Logs

ssh root@apolo 'journalctl -u gohorsejobs-backend-dev -f --no-pager'

---

Troubleshooting

CI/CD Workflow Not Starting

Symptom: "Waiting for runner..."

Solution:

1. Check .forgejo/workflows/deploy.yaml for runs-on label
2. Verify runner has matching label in config

SSH Connection Failed

Solution:

1. Verify public key is in ~/.ssh/authorized_keys
2. Check IP accessibility from runner network

Build Failed in Podman

Solution:

1. Run podman logs -f <container> for details
2. Verify user has podman permissions (rootless or sudo)