How to rebuild a failed RAID array without permanently losing data.
The RAID configurations we will examine have different fault tolerances, ranging from zero to two. Fault tolerance refers to how many hard drives the RAID can lose while staying operational.
First, let’s try and understand what failed in your RAID.
What caused my RAID to fail?
- Hard drive
- Unless your RAID is fault-tolerant, a hard drive failure will leave your entire RAID useless. What causes hard drives to fail?
- RAID 10, 5 and 6 are all capable of functioning in degraded mode and if you are able to replace the failed drives, the RAID should be able to parse together the data from the failed drive(s), even after a hard drive failure.
- RAID 6 is capable of repairing multiple failed drives, while RAID 5 is configured to only repair one drive at a time.
- Controller card
- Controller cards are connected to both the motherboard and your drives. Essentially they control your hard drives within the array.
- Although rare, a controller card failure will be an immediately noticeable problem, slowing down your system considerably.
- The exact same controller card should be used to replace the failed one.
- Do NOT attempt to rebuild RAID if you think your controller is the problem, you could corrupt or lose data.
- Each controller card is different and comes with its own set of capabilities.
- With controller card issues, it is highly recommended to consult a professional data recovery service, as the chance of losing data is much higher.
When do I need a RAID recovery service?
If you know how to rebuild RAIDs confidently, great! If not, talk to an expert. We guarantee to minimize your data loss.
Rebuilding a RAID on your own can be intimidating and for those who are new to the process, you should be certain which condition your RAID is in. If a hard drive has already failed, your steps need to be different than if your RAID has detected a potential failure.
Pre-failure recovery vs Post-failure recovery
- Pre-failure – a RAID can detect a hard drive will fail soon. It will attempt to copy the data from that drive to a spare drive if possible. This is only possible in a configuration that contains a spare. If the spare is used to copy data, and the failing drive does go on to fail, you are now running on a degraded RAID. In this state, the RAID is very prone to failure and data loss. Replacing the failed drive with an identical one is the only way to achieve a healthy RAID again.
- Post-failure – in this instance, your RAID has already failed or is currently running in degraded mode with one or more failed drives.
- If your RAID is still functioning in a degraded mode, you will need to find an identical replacement for your failed one in a timely manner. Time is not your friend here since a hard drive failure while your RAID is in degraded mode could cause catastrophic data loss.
- If your RAID has completely failed and data is inaccessible, please contact us for RAID recovery support. Do NOT attempt to rebuild at this point, you could corrupt and lose ALL your data.
Guidelines for Rebuilding RAID – By Configuration
Be smart and methodical about how you approach your rebuild, making sure to keep everything in the same place it was before you began. Label drives, wires, racks, etc.
Before you run a RAID rebuild, you need to understand the problem.
If you have important data on a failed RAID 0, call us for data recovery service.
If a drive fails in a RAID 1 configuration, that drive’s data is copied onto another disk in the RAID. It will be able to continue running without any data loss.
More than one drive failure at a time will cause RAID 1 to fail. These cases will require a professional data recovery team.
This configuration is tolerant of a single drive failure. In the case of a drive failure, the RAID 5 will attempt to use its parity function, pulling data from throughout the RAID to calculate the missing data.
The more drives and capacity you add, the longer rebuilding takes.
If your RAID 5 fails and you aren’t sure if there has been data loss, contact a professional data recovery service. We can help you avoid data loss in these situations.
Very similar to RAID 5, but is tolerant to more than one hard drive failure. This setup can be complex, and should be rebuilt and monitored with extreme care.
More than two hard drive failures will cause RAID 6 to fail. Data recovery service is required at this stage.
If you have questions about RAID 6 data recovery or rebuilding, don’t hesitate to reach out to one of our RAID experts!
RAID 10 (0+1)
Determine which hard drive has failed, and replace that drive with an identical one. Only one drive in a RAID 10 mirrored set can fail.
Do NOT attempt to access the array (RAID) during the rebuild process. While you may be able to access it, but this could put it at risk of failure.
How long does a RAID take to rebuild?
In general, RAID rebuild time depends on two things:
The quantity of data being calculated and the capacity of the RAID.
Answer: Smaller RAID can rebuild in just a few hours, while more complex ones can take well over 24 hours.
RAID rebuild times are important for storage administrators, especially in enterprise settings, because business can’t be done if data can’t be saved, transferred, etc. Again, some configurations are able to anticipate a hard drive failure and will begin copying data to a spare drive without loss of data.
SALVAGEDATA’s data recovery service experts have completed RAID recoveries for individuals, Fortune 500 companies, and federal institutions. Call our RAID recovery help service line at 800-972-DATA (3282) for immediate assistance – day or night!