Hardware Troubleshooting | Metal on Cloud Help Center

Dedicated server is unreachable

Dedicated server is unreachable Use this when a dedicated server stops responding. Steps 1. Check whether the issue is network, power, OS, or application related. 2. Test from multiple source networks. 3. Request console or remote hands review if OS access is unavailable. 4. Preserve logs after recovery. Useful information to include - Server ID - IP address - Last known good time Escalation Contact Metal on Cloud support when the issue affects production, requires privileged access, involves billing, or needs a formal incident record.

Last updated on May 18, 2026

IPMI or KVM access issue

IPMI or KVM access issue Remote console access may fail because of network, credentials, or browser compatibility. Steps 1. Confirm authorized contact and access window. 2. Try a supported browser and network. 3. Verify VPN or management network requirements. 4. Request credential rotation if access appears stale. Useful information to include - Server ID - User email - Error screenshot Escalation Contact Metal on Cloud support when the issue affects production, requires privileged access, involves billing, or needs a formal incident record.

Last updated on May 18, 2026

Disk failure or RAID alert

Disk failure or RAID alert Disk and RAID alerts need quick evidence and replacement planning. Steps 1. Capture RAID status and disk serial if available. 2. Confirm backup status before replacement. 3. Schedule replacement or emergency action. 4. Validate rebuild and service health after work. Useful information to include - Disk serial - RAID state - Backup status Escalation Contact Metal on Cloud support when the issue affects production, requires privileged access, involves billing, or needs a formal incident record.

Last updated on May 18, 2026

High temperature or power alert

High temperature or power alert Thermal and power alerts require facility and hardware checks. Steps 1. Record sensor values and timestamps. 2. Check whether neighboring devices show similar alerts. 3. Request facility airflow or power review. 4. Plan hardware inspection if alerts continue. Useful information to include - Sensor readings - Rack location - Time window Escalation Contact Metal on Cloud support when the issue affects production, requires privileged access, involves billing, or needs a formal incident record.

Last updated on May 18, 2026

Unexpected reboot analysis

Unexpected reboot analysis Unexpected reboots should be treated as incidents until cause is known. Steps 1. Collect OS logs and uptime history. 2. Check power, hardware, and kernel events. 3. Review maintenance notices. 4. Request facility or hardware investigation if unclear. Useful information to include - Server ID - Reboot time - OS logs - Application impact Escalation Contact Metal on Cloud support when the issue affects production, requires privileged access, involves billing, or needs a formal incident record.

Last updated on May 18, 2026