Despite all the data loss prevention measures taken, it can still happen that a server fails or a storage system loses its data consistency. Data and storage redundancies do not always guarantee the expected reliability.
Data recovery from RAID or NAS systems not only requires extensive knowledge of the individual configuration options of the array or drive network, but also very broad and professional expert knowledge of the individual file systems that can be used in such systems. Our service for data recovery from failed NAS / SAN and server systems not only deals exclusively with the logical, but also with the physical rescue of defective hard drives in a RAID network / array.
Considering the situation, it is advisable to refrain from any attempt at data recovery on a RAID or NAS / SAN and to contact a professional data recovery company. Here you will receive the appropriate advice and information on the best course of action to best maintain the chances of restoring your redundant drive network.
Restoring lost data or rebuilding RAID systems should generally only be attempted on sector-based copies of each individual drive as part of the network. This is the only way to ensure that you do not inadvertently falsify the original condition after such a system fails or influence it so significantly that it can no longer be rescued by a laboratory. Every write access is a physical intervention in the file structure of your RAID, which in the worst case can lead to irreversible damage.
Due to the variety of RAID types (hardware RAID or software RAID) and levels (e.g. RAID 0, RAID 1, RAID 5, RAID 6) as well as different degrees of virtualization, the possible failure scenarios are diverse.
Typical failure symptoms are:
1) RAID is offline, rebuild aborted after another hard drive fails
What should you do if another disk suddenly fails during a rebuild? This can be caused by defects in individual hard drives and subsequent errors in the RAID configuration.
If a hard drive in the RAID array fails, redundancy is lost with common RAID levels and in most cases this results in serious damage.
If an error message such as drive missing or disk failed appears with RAID-5, RAID-6, RAID-10 and similar RAID levels, due to the redundancy of these systems, it is possible to replace the defective hard drive with a new one and perform a rebuild to start. Depending on the configuration, this process can be done with or without a hot spare hard drive. During a rebuild, the RAID array uses certain RAID algorithms and parity data to copy the data from the defective hard drive to a new drive or to a hot spare disk. During this process, all hard drives in the RAID array are put under a lot of strain, so it is not uncommon for one or even several additional hard drives to fail during the rebuild.
2) RAID / NAS failure after server restart
RAID technology is intended to increase the reliability and performance of hard drive systems. But data loss can also happen with a RAID system. Error messages such as RAID offline, RAID critical, RAID SUB-OPTIMAL, drive missing, disk failed are not uncommon and indicate a serious problem.
The cause can be one or more defective hard drives in the RAID or NAS. It is also possible that the RAID controller fails.
In some cases and with smaller RAID configurations or NAS with a single hard drive, no access is possible at all and only the status LED of the NAS flashes red/green alternately.
Under no circumstances should you press the reset button on the NAS housing in this case; this is used to reset the NAS to factory settings in order to reinstall the operating system. The data is lost.
3) Voltage spikes, power outage
After a power failure or overvoltage, a short circuit is just one of the many possible consequences.
Modern systems are actually well protected against overvoltage, especially with NAS / RAID systems, the CPU or UPS devices ensure that the hardware is not damaged.
If possible, a separate circuit should be used for NAS and RAID arrays.
But it often happens that problems arise after a voltage spike or power failure. In such a scenario, it always depends on the “unfavorable” time at which it happens.
If the system was in operation during this time, serious mechanical and logical defects could occur in the array or NAS. After restarting the system, the IP address, for example, is not assigned correctly, or the NAS is no longer found. With RAID systems, the configuration can also be damaged by an abrupt shutdown. Even the classic, lightning strike often means that failures of drives and other components do not occur independently of one another. The overvoltage caused by a lightning strike suddenly increases the probability of failure of the entire network.
4) Defective RAID controller in the server (or failure after firmware update)
A firmware update of the RAID controller and the installation of new drivers are not always enough to get the RAID system back into operation after such an error. After the typical reboot, the status LED of the NAS flashes red/green alternately.
If the controller fails completely or the file system is corrupted by a controller, without this being immediately noticed, access to the start partition is then limited and the data partition cannot be viewed at all.
A hardware RAID controller is a physical controller that is built into the server as a hardware component. This controller has its own processor that organizes and manages the storage space to calculate the RAID operations.
Simply replacing the RAID controller without losing data is therefore impossible. The new controller will not be able to access the contents of the previous RAID array.
QNAP NAS servers are a commonly used storage solution. New firmware updates are regularly offered by the manufacturer. Even if the firmware update is intended to lead to improved system properties by fixing any existing errors (bugs) and adding new functions for RAID and NAS, it can happen that an update fails.
The causes can be different: A missing software package or utility that should have been installed additionally.
Sometimes following the manufacturer’s instructions is not enough to troubleshoot errors. In the worst case, the device will even be reset.
5) After running chkdsk, fsck, the data is no longer accessible
On Windows based systems the CHKDSK and on Linux based systems the FSCK is a tool to check the integrity of the file system and troubleshoot errors. CHKDSK / FSCK checks and repairs lost allocation units (clusters) as well as incorrect directory entries.
Under certain circumstances, the file system of a RAID or NAS may be corrupted, restricting access to the data. This can cause warning messages to appear or cause the system to behave “strangely,” such as processes crashing and software not starting.
If errors cannot be corrected, for example due to defective sectors on individual hard drives, the result is often damage to the file system structure and the logical integrity of a file system is no longer guaranteed.
It is therefore important to avoid these file system checks in the event of error messages, as this can result in logical damage to the network when the storage is subsequently restarted.
6) RAID configuration is lost or data and directories are deleted
Data losses caused by incorrect initialization of the RAID system, lost configurations or incorrect parity calculation are not uncommon. Most errors occur during RAID management processes.
The systems offer countless possibilities, e.g. expanding capacity by adding additional hard drives, spares or regularly replacing the built-in data storage media.
With NAS systems, permissions can be changed for individual or multiple user groups. All of these operations do not occur automatically, but are carried out by users or administrators. Not everything always goes according to plan.
After restarting, “no operating system found” suddenly appears. An incorrect RAID level was set or data, directories or shares were accidentally deleted. These are just a few of the possible scenarios.
Correct behavior is crucial for all logical errors in RAID systems. Since these are complex systems, any self-attempt can have serious consequences. Further corruption of the file system structure can destroy the logical integrity of a file system. In addition, operating systems such as Linux release all deleted “inodes”, i.e. file system entries, for immediate use
Which manufacturers do we save and reconstruct the data from?
We rescue RAID and NAS configurations from all manufacturers and have decades of expertise in this area. This experience includes not only the individual levels of the respective RAID, but also the different file systems used.
The manufacturers of RAID / NAS and server systems where we are often consulted for restoration and repair include:
- Buffalo Technologies (Linkstation / Terastation, etc.)
- Western Digital
- Seagate
- Dell / EMC
- Fujitsu
- Hitachi
- HP
- IBM
- Netgear
- Netapp
- Qnap
- Synology and Zyxel
- and many more
Before we start rescuing your RAID or NAS system, we first carry out an analysis for determining the cause and level of damage. We determine the effort required to save your data and send you your individual offer by email.