# Support ticket evidence collection

Source: https://docs.quake.ai/docs/operate/troubleshooting/support-ticket-evidence
Markdown: https://docs.quake.ai/docs/operate/troubleshooting/support-ticket-evidence.md
> What to collect before filing a support ticket: commands, logs, and identifiers for faster resolution.

---

# Support ticket evidence collection

When a problem is not self-resolvable and you need to escalate to Quake AI support, collecting the right evidence upfront reduces back-and-forth and speeds up resolution. This page provides a reusable checklist that works across compute, network, storage, and Kubernetes incidents.

## Before you start

- Have the OpenStack CLI installed and authenticated (`source openrc.sh`)
- Note the **UTC timestamp** when the problem first occurred; this is the most important piece of context for the support team
- If possible, keep the affected resources in their current state until you have gathered evidence; do not delete or recreate resources before capturing diagnostics

## General evidence checklist

Collect these items regardless of incident type:

| Item | Command | Why it matters |
|---|---|---|
| Project ID | `openstack project show YOUR_PROJECT -c id` | Identifies your project in platform logs |
| Timestamp (UTC) | Record when the problem started | Allows support to correlate with platform events |
| Quota usage | `openstack quota show --usage` | Rules out resource exhaustion |
| Token validity | `openstack token issue` | Confirms your CLI session is authenticated |
| Recent changes | Note any security group, network, or configuration changes made before the issue | Identifies potential self-inflicted causes |

## Compute evidence

| Item | Command |
|---|---|
| Instance ID and status | `openstack server show YOUR_INSTANCE -c id -c status -c fault` |
| Console log (last 50 lines) | `openstack console log show YOUR_INSTANCE --lines 50` |
| Instance details | `openstack server show YOUR_INSTANCE` |
| Key pair used | `openstack server show YOUR_INSTANCE -c key_name` |

## Network evidence

| Item | Command |
|---|---|
| Network ports | `openstack port list --server YOUR_INSTANCE` |
| Port details | `openstack port show YOUR_PORT -c security_group_ids -c fixed_ips -c status` |
| Security group rules | `openstack security group rule list YOUR_SECURITY_GROUP` |
| Floating IP details | `openstack floating ip show YOUR_FLOATING_IP` |
| Router details | `openstack router show YOUR_ROUTER` |

## Block storage evidence

| Item | Command |
|---|---|
| Volume ID and status | `openstack volume show YOUR_VOLUME -c id -c status -c attachments` |
| Snapshot status | `openstack volume snapshot show YOUR_SNAPSHOT -c id -c status` |
| Volume list | `openstack volume list` |

## Object storage evidence

| Item | How to get it |
|---|---|
| Access key (not the secret) | `openstack ec2 credentials list -c Access` |
| Endpoint URL used | The value of `AWS_ENDPOINT_URL` or `--endpoint-url` in your client configuration |
| Full error response | Copy the complete error message or XML/JSON response from your client |
| Request details | Note the HTTP method, URL, and any special headers used |

## Kubernetes evidence

| Item | Command |
|---|---|
| Cluster ID and status | `openstack coe cluster show YOUR_CLUSTER -c uuid -c status -c status_reason` |
| Heat stack ID and status | `openstack stack show STACK_ID -c stack_status -c stack_status_reason` |
| Failed stack resources | `openstack stack resource list STACK_ID --filter status=FAILED` |
| Master instance console log | `openstack console log show MASTER_INSTANCE --lines 100` |
| kubectl output (if cluster is reachable) | `kubectl get nodes -o wide` and `kubectl cluster-info` |

## Data hygiene before sharing

Before including evidence in a support ticket, redact or remove:

- **Passwords and secrets**: never include passwords, private keys, or secret keys in a ticket
- **EC2 secret keys**: share only the access key, not the secret key
- **Application credential secrets**: share only the credential name or ID
- **Personal data**: redact usernames, email addresses, or other personal information that is not relevant to the issue
- **Environment variables**: if sharing shell output, verify that no credentials leaked into the output


The `openrc.sh` file contains your application credential secret. Do not attach it to a support ticket. If support needs to verify your credential scope, share only the credential name or ID.


## Filing the ticket

Include in your support ticket:

1. **One-sentence summary** of the problem
2. **UTC timestamp** when it started
3. **Affected resource IDs** (instance, volume, cluster, network, as applicable)
4. **What you already tried**: steps taken and their results
5. **Evidence** from the checklists above
6. **Reproduction steps** if the issue is repeatable

File tickets through the [Quake AI support portal](https://portal.rumble.cloud/help).

## See also

- [Troubleshooting overview](/docs/operate/troubleshooting)
- [Auth token diagnostics](/docs/operate/troubleshooting/auth-token-diagnostics): 401/403 errors, token expiry, credential scope
- [Quota and limits troubleshooting](/docs/operate/troubleshooting/quota-and-limits): quota exhaustion across compute, network, and storage
- [Instance connectivity troubleshooting](/docs/operate/runbooks/instance-connectivity)
- [Instance lifecycle troubleshooting](/docs/operate/runbooks/instance-lifecycle)
- [Volume troubleshooting](/docs/operate/runbooks/volume-troubleshooting)
- [Object storage access troubleshooting](/docs/operate/runbooks/object-storage-access)
- [Kubernetes cluster troubleshooting](/docs/operate/runbooks/kubernetes-troubleshooting)