> ## Documentation Index
> Fetch the complete documentation index at: https://docs.cyberup24.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Escalation Criteria

Most issues with the LARK can be resolved on-site using the guidance in the troubleshooting section. When an issue falls outside what on-site troubleshooting can resolve, or when the situation puts data integrity or hardware at risk, escalating to support is the right next step. This page describes the conditions that warrant escalation and what to gather before you make contact.

## Escalate immediately

Contact support without waiting to exhaust on-site options when any of the following are true:

* **Hardware damage is suspected.** If the appliance has been dropped, flooded, or exposed to conditions outside its operating range, stop using it and contact support before attempting a restart.
* **A hard drive failure is suspected.** Continued operation with a failing drive can worsen data loss. See [Responding to Hard Drive Failure](/troubleshooting/hard-drive-failure).
* **POST or BIOS errors cannot be cleared.** If the appliance will not complete POST after reseating the management module and cycling power, the failure is likely a hardware fault. See [BIOS / POST Failures](/troubleshooting/bios-post-failures).
* **Data integrity is at risk.** If storage errors, incomplete writes, or corrupted captures are observed and the condition is not explained by a power loss event with a clear recovery path.

## Escalate after on-site troubleshooting

Contact support after attempting the relevant on-site troubleshooting steps when:

* An issue recurs after following the documented resolution steps
* Pod errors persist for more than 30 minutes and log review does not reveal a cause. See [Pod Errors](/troubleshooting/pod-errors)
* Performance degradation does not improve after addressing storage and resource contention
* Keycloak admin lockout cannot be resolved using the recovery procedure. See [Keycloak Admin Lockout](/troubleshooting/keycloak-admin-lockout)
* An issue is not covered by any troubleshooting page in this documentation

## What to gather before contacting support

Providing the following information when you contact support allows the team to respond faster and more accurately:

| Item                                   | Where to find it                             |
| -------------------------------------- | -------------------------------------------- |
| LARK model (SWIFT, PEREGRINE, or APEX) | Delivery record                              |
| Serial number                          | Delivery record or appliance label           |
| Software version                       | Admin interface → System status or About     |
| Description of the symptom             | Your observation                             |
| Steps already taken                    | What you tried and the results               |
| Error messages or codes                | Admin UI, system status view, or BMC console |
| Screenshots of error states            | System status page, pod error view           |
| Approximate time the issue began       | Your records                                 |

<Note>
  Contact details for support are provided on the delivery record that accompanies the appliance, or were communicated separately at the time of purchase. If you cannot locate your escalation contact information, reach out through the channel your organization uses to manage the LARK deployment.
</Note>

<Tip>
  When capturing screenshots of error states, include the full system status page and any individual pod or service detail views that show the error. More information is always better than less when diagnosing remotely.
</Tip>