A server that stops booting right after someone replaced, added or moved a drive usually still has its data. In Dell and HPE documentation the usual causes are a controller that is waiting for a decision about a foreign or missing configuration, drives that are missing or not seated, a wrong boot mode or boot order, or a boot volume setting on the controller. This guide walks through those checks in a safe order for PowerEdge servers with PERC controllers and ProLiant servers with HPE SR (Smart Array) or MR controllers.
Before you start
- Write down exactly what changed: which slot, which drive was removed, what was inserted, and whether the server was on or off.
- Have iDRAC9 or iLO access so you can read the logs and use the remote console.
- Do not move more drives around while you investigate. HPE drive roaming, for example, is an offline feature and requires the logical drive status to be good before drives are moved.
- Confirm when the last good backup was taken.
Warning: Do not clear a foreign configuration, initialise, delete or recreate a virtual disk (logical drive) to "fix" a boot problem. Dell's PERC 11 guide notes that clearing a foreign configuration returns the disk to Ready state, and that even a fast initialisation overwrites the first and last 8 MB of the virtual disk, clearing boot records and partition information. Back up first, and only change the RAID configuration once you know the cause.
Step 1 — Read the POST message and the logs
The message on the console tells you which path to follow. Open the remote console and read it before pressing any key.
- Dell: in iDRAC9 open Maintenance > System Event Log and Maintenance > Lifecycle Log, then Storage > Overview > Summary, Virtual Disks and Physical Disks. See reading the SEL and Lifecycle Log.
- HPE: in iLO open Information > Integrated Management Log and System Information > Storage. See using the IML. On Gen11, if the server will not start, the UID button can show the iLO Server Health Summary (with the most recent critical IML events) on a connected monitor, provided Show Server Health on External Monitor is enabled in iLO. Press and release it; HPE warns that holding it for more than five seconds starts an iLO reboot.
| Message | What the vendor says to do |
|---|---|
| PERC: "Foreign configuration(s) found on adapter. Press any key to continue, or 'C' to load the configuration utility or 'F' to import foreign configuration(s) and continue." | Press F to import only if all member disks of the virtual disk are present, or press C to enter the HII utility and review it. |
| PERC: "The following virtual disks are missing ..." | Check cable connections, restart, then import the virtual disk or discard the preserved cache in HII. The same screen offers X to permanently discard cached data; do not press it unless you have decided the data is not needed. |
| PERC: "Entering the configuration utility in this state will result in drive configuration changes ..." | Check cables first; continue only if there are no cable problems. |
| HPE: "Slot %1 Storage: Drive(s) at %2 are missing from volume" | Power down, check cables, make sure all drives are fully seated, power cycle external enclosures while the system is off, then power up. |
| HPE: "Drive Array - New (or Previously Failed) Logical Drive(s) Attachment Detected" | Informational: drives were connected while power was off. No action required. |
For importing or clearing on PERC, follow import or clear a PERC foreign configuration.
Step 2 — Confirm the right drive was changed and all members are present
- If several drives of one virtual disk became inaccessible at once, Dell's PERC 11 procedure is: turn off the system, check cable connections and reseat the physical disks, make sure all disks are present, turn on, and import the foreign configuration. A virtual disk that went directly Offline because of a cable pull or power loss is imported as Optimal without a rebuild; one that was Degraded first starts a rebuild after the import.
- On PERC, a drive removed and reinserted in the same slot shows as foreign for a short time before rebuilding. Dell says no action is required.
- On HPE Smart Array, if logical drives disappear after drives were moved to another server or enclosure, HPE says to follow the drive roaming rules: drive roaming is an offline feature (there is no method for moving an array while the server is online), and the logical drive status must be good before drives are moved.
Step 3 — Check boot mode and boot order
If the virtual disk or logical drive is healthy but the server still does not find the OS, check the system firmware.
- Dell: press F2 and open System Setup Main Menu > System BIOS > Boot Settings. Boot Mode is UEFI by default, and Dell cautions that switching the boot mode may prevent booting if the OS was not installed in the same mode. Change the order under System BIOS > Boot Settings > UEFI Boot Settings > UEFI Boot Sequence using the + and - keys.
- HPE: in System Utilities open System Configuration > BIOS/Platform Configuration (RBSU) > Boot Options > UEFI Boot Settings > UEFI Boot Order and move the OS entry up. The Gen10/Gen10 Plus UEFI guide also documents Boot Options > Boot Mode (UEFI Mode by default, or Legacy BIOS Mode) and states that the boot mode must match the operating system installation.
Step 4 — Check the controller boot device (legacy BIOS mode only)
Both vendors document a controller-level boot volume setting that applies only to legacy BIOS boot mode. In UEFI mode the system boot order from Step 3 is what matters.
- Dell PERC 11: System Setup > Device Settings > (your PERC) > Main Menu > Controller Management > Select Boot Device. If no boot device is selected, the first virtual disk becomes the boot device on the next reboot. The iDRAC RACADM guide documents setting it remotely:
racadm storage setbootvd:<Controller FQDD> -vd <VD FQDD>- HPE SR controllers: the SSA CLI guide documents the boot controller and boot volume for legacy BIOS boot mode, and HPE recommends keeping the default (the controller boots from the first logical drive) to avoid boot errors after replacing the boot volume or controller. Syntax from the guide, with its example slot and drive numbers:
controller slot=1 modify bootcontroller=enable
controller slot=1 ld 1 modify bootvolume=primaryThe HPE Gen11 troubleshooting guide lists "the controller boot order is not set properly" and "a logical drive is not configured" among the causes of a server that does not boot, and says to verify the controller boot order in the UEFI System Utilities.
Step 5 — When to stop and contact the vendor
- PERC reports "Firmware is in Fault State", or fatal errors or data corruption when accessing virtual disks. Dell's corrective action for both is to contact Global Technical Support.
- More drives are missing or failed than the RAID level can tolerate. HPE describes this as compromised fault tolerance, where data loss can occur. See virtual disk degraded or offline.
- The controller itself is not visible during POST after cable and seating checks.
Before calling, collect a SupportAssist collection (Dell) or an Active Health System log (HPE) so support can see the controller history.
Frequently asked questions
Should I clear the foreign configuration to get the server booting?
No. Import it only when all member disks are present. Clearing removes the configuration from those disks and returns them to Ready state.
Why does the server not boot after I changed the boot mode?
Dell and HPE both state that the boot mode must match the mode the operating system was installed in. Set it back to the original mode.
Checklist
- POST message read and recorded; SEL, Lifecycle Log or IML checked.
- Foreign configuration imported only with all member disks present; nothing cleared or initialised.
- Cables and drive seating checked with the server powered off.
- Boot mode matches the OS installation; OS entry is first in the UEFI boot order.
- Controller boot device checked if the server uses legacy BIOS mode.
- Logs collected and a vendor case opened if the controller faulted or volumes failed.
Skyline can supply, configure and maintain Dell, HPE and Huawei servers under an annual maintenance contract (AMC). Contact us.
Comments
0 total · 0 threads