# Reducing Failed Releases for Online Poker

> TechSupport Team supported the deployment pipeline of RiverCard’s C++ and Erlang poker platform. Over four months, the client’s GitLab CI records show failed releases falling from 24% to 9%. Release approval stayed with the client’s engineering lead.

- Publisher: TechSupport Team
- Canonical page: https://techsupport-team.com/case-studies/online-poker-release-reliability/
- Updated: 2026-11-16

Client: RiverCard · iGaming & Betting · Gibraltar online poker network

## Client and engagement facts

- **Client:** RiverCard
- **Industry:** iGaming & Betting
- **Business context:** Gibraltar online poker network
- **Delivery model:** DevOps support
- **Application:** Poker platform with tournament scheduler
- **Supplied geography:** Gibraltar

## Application stack and tools

### Application stack

- C++/Erlang
- Docker
- Kubernetes
- Terraform

### Tools used

- GitLab CI
- Prometheus
- Argo CD

Related services: [DevOps support](https://techsupport-team.com/services/devops-support/)

## Why Support Was Needed

A scheduler release interrupted play for 3,200 users. Deployment instructions lived in an unmaintained wiki, and engineers lacked a consistent first check when a release failed.

## Technical Work

- Converted wiki instructions into deployment runbooks with ordered first checks.
- Validated the runbooks in a production-like staging cluster for one week.
- Added pre-deployment checks for container image tags and Kubernetes ConfigMaps.

## Representative Incident

The first release after the pipeline migration failed on a hard-coded image tag. The team reproduced the failure in staging and added the condition to the pre-deployment checklist. The next ten deployments completed without rollback.

## Results and how they were measured

Table: Reported results and their measurement scope

| Metric | Baseline | Result | Units | Measurement window | Source and scope |
| --- | --- | --- | --- | --- | --- |
| Failed releases | 24% | 9% | % | 4 months | Client GitLab CI; Scope: One release failure during a cloud-region outage was excluded. |
| MTTR | 52 min | 16 min | min | 4 months | Client Prometheus |
| Deploys/week | 2 | 9 |  | 4 months | Client ArgoCD |

## Evidence and Limitations

The comparison covers four months. One release failure during a cloud-region outage was excluded.

## Release Approval

TechSupport Team prepared and tested the deployment runbooks. RiverCard’s engineering lead approved each production release.

## Scope a Similar Engagement

Share a recent release failure, your CI/CD setup and who approves production changes. [Request a support plan](https://techsupport-team.com/contact/) or explore [DevOps support](https://techsupport-team.com/services/devops-support/).

## Related pages

- [DevOps support](https://techsupport-team.com/services/devops-support/): The related service for this engagement.

## Contact

Request engineers from TechSupport Team or ask a question: https://techsupport-team.com/contact/
