Skip to content

Getting Help

When working with Python Discord's infrastructure, you may encounter issues or need assistance. This guide outlines the various ways to get help and who to contact for different types of problems.

🆘 Emergency Procedures

Critical Infrastructure Issues

For critical incidents affecting production services:

  1. Immediate Response: Post in the #dev-oops Discord channel with @here mention
  2. Assessment: Briefly describe the issue and impact
  3. Escalation: If no response within 15 minutes, contact DevOps team leads directly

After-Hours Emergencies

For emergencies outside normal hours:

  1. Contact the on-call DevOps team member (rotation schedule in #dev-oops pinned messages)
  2. If unable to reach on-call person, escalate to DevOps team leads
  3. Document the incident for post-mortem analysis

🙋 Getting Help by Problem Type

Infrastructure Questions

For questions about services, configurations, or deployments:

Service-Specific Issues

For problems with specific services:

Service Type Primary Contact Resources
Kubernetes Cluster #dev-oops channel Common Queries
PostgreSQL Database #dev-oops channel PostgreSQL Queries
Email Services #dev-oops channel Email Documentation
LDAP/Authentication #dev-oops channel LDAP Documentation
DNS Issues #dev-oops channel DNS configuration in infra repo
Monitoring/Alerting #dev-oops channel Loki Queries

Access and Permissions

For access issues or permission requests:

  1. Check the Access Table for current permissions
  2. Request access in #dev-oops channel with:
  3. What access you need
  4. Why you need it
  5. How long you need it (if temporary)

📚 Self-Help Resources

Before asking for help, check these resources:

Documentation Sections

Quick Troubleshooting

  1. Check service status: Look at monitoring dashboards
  2. Review recent changes: Check recent deployments or configuration changes
  3. Search past incidents: Look through post-mortems for similar issues
  4. Verify access: Ensure you have the necessary permissions

🧰 Troubleshooting Workflows

Service Down or Unresponsive

  1. Immediate: Check if it's a known incident (#dev-oops announcements)
  2. Investigate:
  3. Check Kubernetes pod status
  4. Review application logs
  5. Verify dependencies (database, external services)
  6. Escalate: If unable to resolve in 30 minutes, ask for help

Performance Issues

  1. Gather data: Collect metrics and logs showing the performance problem
  2. Check resources: Review CPU, memory, and network usage
  3. Recent changes: Identify any recent deployments or configuration changes
  4. Document: Include specific metrics when asking for help

Configuration Problems

  1. Verify syntax: Check YAML/configuration file syntax
  2. Compare with working examples: Look at similar working configurations
  3. Test in staging: If available, test changes in non-production environment
  4. Rollback plan: Have a rollback strategy before making changes

:phone: Contact Information

Discord Channels

  • #dev-oops: Primary channel for all infrastructure discussions
  • #admin-chat: For administrative and governance discussions

GitHub

When to Use Each Channel

Type of Request Discord #dev-oops GitHub Issue
Urgent issues ✅
Quick questions ✅
Bug reports ✅ ✅
Feature requests ✅
Documentation improvements ✅
Complex discussions ✅

📝 When Asking for Help

Information to Include

Always provide:

  • What you're trying to do: Clear description of your goal
  • What you expected: What should have happened
  • What actually happened: What went wrong (include error messages)
  • Environment: Which service/system is affected
  • Recent changes: Any recent modifications that might be related

For service issues, also include:

  • Timestamps of when the issue started
  • Affected services or users
  • Current impact level
  • Steps already taken to troubleshoot

Example Help Request

🚨 PostgreSQL Connection Issues

**Goal**: Deploy new bot update to production
**Expected**: Bot should connect to database normally
**Actual**: Getting connection timeout errors

**Environment**: Production bot deployment
**Started**: ~2:30 PM UTC
**Impact**: Bot is offline, affecting all Discord functionality

**Error**: `connection to server at "postgres.pydis.svc.cluster.local" (10.2.3.4), port 5432 failed: timeout expired`

**Already tried**:
- Checked pod logs (show connection attempts)
- Verified database pod is running
- No recent config changes

**Need help**: Investigating why connections are timing out

📚 Learning Resources

For New Team Members

  1. Start with Onboarding documentation
  2. Review DevOps Rules
  3. Explore Available Tools
  4. Shadow experienced team members during incidents

Skill Development

  • Kubernetes: Practice with local clusters, review runbooks
  • PostgreSQL: Study common queries and maintenance procedures
  • Monitoring: Learn Grafana, Prometheus, and Loki query languages
  • Infrastructure as Code: Understand our Ansible and Kubernetes manifests

❓ Frequently Asked Questions

"I'm new to the team, where do I start?"

  1. Read the Onboarding guide
  2. Get required access from the Access Table
  3. Join the #dev-oops Discord channel
  4. Introduce yourself and ask for a team overview

"I made a mistake, what should I do?"

  1. Don't panic: Mistakes happen and are learning opportunities
  2. Assess impact: Determine if it's affecting production services
  3. Communicate: Post in #dev-oops immediately if there's any impact
  4. Document: Help create a post-mortem to prevent future occurrences

"I'm not sure if this is urgent, should I ask for help?"

When in doubt, ask! It's better to ask for help unnecessarily than to let a small issue become a major incident.

Contributing to This Guide

Found something unclear or have suggestions for improvement?

  • Open an issue on the infra repository
  • Suggest changes via pull request
  • Bring it up in #dev-oops for discussion

Remember: Good documentation helps everyone work more effectively! 🚀