Emergency recovery for SCO servers
The server will not boot, a disk has failed, or nobody can log in. Send the details now. Recovery work is available outside UK office hours.
Before you do anything else
- Stop restarting a server with a failing disk. Every start risks more damage.
- Do not reinstall over the existing disk.
- Photograph any error on the screen.
Report a server that is down
Say what happened, what the screen shows and whether you have a backup.
The failures we are asked about most
It will not boot
Boot errors, kernel panics, a damaged root filesystem, or a server that stops part-way through start-up.
A disk or controller has failed
Failing IDE or SCSI disks, a dead RAID controller, or a mirror that turned out not to be mirroring.
The hardware has died
No power, a failed motherboard, and no replacement parts. The disks are usually still readable.
Nobody can log in
A lost root password, locked accounts, or the one person who knew the system has left.
Licence or activation errors
The system has stopped accepting users or reports an expired licence, often after a hardware change.
The backup will not restore
An unreadable tape, a backup that has silently failed for months, or no record of how to restore it.
Copy first, then repair
The original disks are the only copy of your system. We protect them before attempting any repair, so that a failed attempt costs time and nothing else.
Recovered systems are usually brought back as a virtual machine. That gets the business working again without waiting for parts that may no longer exist, and leaves you on hardware that will not fail the same way twice.
-
Triage
What failed, what still works, what backups exist, and how long the business can run without the system.
-
Image the disks
A complete copy of every readable disk, taken before anything is changed.
-
Rebuild from the copy
The system is repaired and started from the image, usually on a virtual machine.
-
Check it with you
Users log in, documents print, and the figures match before anyone relies on it.
-
Make sure it cannot happen again
Working backups, a note of what the system is, and a plan for the hardware.
The details that save the most time
- What happened, and when it last worked
- The SCO release, if you know it, and the make and model of the server
- The exact wording of any error, or a photograph of the screen
- Whether you have a backup, what kind, and how old it is
- What the system runs, and what the business cannot do without it
Send it through the form on this page or to help@scoexpert.com, with "server down" at the start of the message.
OpenServer 5.0.5 on an HP NetServer E60
A business system on hardware from the 1990s, restored and brought back to life as a virtual machine.
Read the rescue storyEmergency recovery questions
How quickly can you start?
Send the details and say that the server is down. Emergency work is available outside normal UK office hours. We will reply with what we can do and when.
Can you recover a server when there is no working backup?
Often, yes. If the disks can still be read, we take an image of them first and work on the copy. That image is then used to rebuild the system, usually as a virtual machine.
What does emergency work cost?
Emergency and out-of-hours work is charged at the emergency rate shown on the pricing page. We tell you the likely cost before starting.
We have lost the root password. Can you get us back in?
Yes, provided you have physical or console access to the server. It is done by starting the system from emergency boot media and resetting the password on the root filesystem. Our guide shows the steps.
Is a server down right now?
Send what you know. The release, the hardware, what the screen says and whether there is a backup.
- You deal with the people who do the work
- No obligation, and no upgrade required before we will help
- We tell you plainly if something is not worth doing