BREAKING: GitHub SSH Deploy Keys Failing — Immediate Workarounds [CRITICAL]

GitHub experiencing widespread SSH deploy key failures. Affecting CI/CD pipelines and automated deployments. Here's what's down, who's impacted, and how to fix it now.

BREAKING: GitHub SSH Deploy Keys Failing — Critical Incident

Status: Investigating | Severity: CRITICAL | Last Updated: NOW

---

What's Down & Who's Affected

GitHub is currently experiencing widespread failures with SSH connections using deploy keys. This primarily impacts:

Automated deployments via CI/CD pipelines (GitHub Actions, Jenkins, CircleCI, etc.)

Server-to-GitHub authentication for pull/push operations

Bot accounts and service integrations relying on SSH-based deploy keys

Infrastructure automation (Terraform, Ansible, etc. pulling private repos)

Unaffected: HTTPS connections, personal SSH keys, and web-based Git operations appear to be functioning normally.

If your deploy pipeline suddenly broke in the last 2 hours, you're affected.

---

Immediate Workarounds (DO THIS NOW)

Option 1: Switch to HTTPS + PAT (Fastest)

1. Generate a GitHub Personal Access Token (PAT): - Settings → Developer settings → Personal access tokens → Tokens (classic) - Scope: repo and workflow

2. Update your deploy config to use HTTPS: ```bash git clone https://USERNAME:TOKEN@github.com/org/repo.git ```

3. Store token securely in your CI/CD secrets (never commit it)

Time to deploy: 2-5 minutes

Option 2: Use Fine-Grained PATs

More secure than classic tokens

Settings → Developer settings → Fine-grained tokens

Grant repository-specific access only

Recommended for new setups

Option 3: Wait for SSH Fix + Fallback

While investigating, configure your CI/CD to retry SSH after 5 minutes:

```yaml

GitHub Actions example

name: Checkout

uses: actions/checkout@v4 with: fetch-depth: 0 ```

Actions should automatically retry SSH connections. If stuck, switch to Option 1.

---

How to Check If You're Affected

Quick Test

```bash ssh -T git@github.com

Should return: Hi USERNAME! You've successfully authenticated...

Test deploy key specifically

ssh -i ~/.ssh/deploy_key -T git@github.com

If this fails = you're affected

```

Check Your Logs

GitHub Actions: Check failed workflow runs for SSH timeout/auth errors

Jenkins/CircleCI: Look for Permission denied (publickey) or Connection timeout

Manual deployments: Run ssh -vvv -i deploy_key git@github.com for debugging

---

Alternative Tools to Consider

If this outage reveals deployment fragility:

GitLab CI/CD — Native alternative with similar workflows

Gitea — Self-hosted Git option

Forgejo — Community fork of Gitea

AWS CodeCommit — For AWS-native teams

Bitbucket — Atlassian's GitHub alternative

*Note: Don't panic-migrate. This is a temporary outage. But it's worth reviewing your deployment resilience.*

---

Monitor Recovery

1. Official Status: https://www.githubstatus.com 2. Twitter: @githubstatus 3. Your Test: ```bash watch -n 30 'ssh -i ~/.ssh/deploy_key -T git@github.com 2>&1' ``` Once you see successful auth → SSH is fixed

4. Revert to SSH once GitHub confirms resolution (usually within hours)

---

Key Takeaway

HTTPS + PAT is your immediate escape hatch. It takes 5 minutes and works reliably. Deploy that now while SSH is being fixed.

GitHub's investigating. Most outages resolve within 2-4 hours. Stay calm, switch protocols, monitor recovery.

—StillNotAThing Dev Team