Does RAID5 mode array fail to recover data? Can it be recovered absolutely? Really damaged 1 disk case
DiskGenius Data Recovery Software Professional Edition supports virtual reorganization of RAID, virtual RAID set up successfully, you can manipulate the disk in the software, such as recovering data, backup files, etc. If the original RAID partition table and file system is not damaged, the original RAID files will be used to recover the data. If the original RAID partition table and file system has not been destroyed, the original RAID files will be in this.
Of course, RAID5
So can the server lost data still be recovered? It can! However, server data loss is generally a more complex issue, because the user actually has to perform RAID-related data recovery operations, so the operator needs to know certain data recovery techniques and have relevant experience. The first thing to remember is that during the data recovery period, do not do any unsure operation to prevent the data that could have been recovered from being destroyed.
A friendly reminder: after the server data loss, do not try to forceonline, rebuild, initialize, reconfigure the disk information, etc., because these operations may lead to the data can never be recovered.
Server data repair, in order to be able to improve the hard disk array data recovery, server administrators (not data recovery professionals) to do?
1, avoid dangerous operations: do not force online
In the server of a hard disk failure, the general server will have an Offline indicator prompt, at this time, the user is required to replace an intact the same hard disk, and recalibrate the algorithm. This is a very normal procedure, but it is not known that many RAID disasters are caused by this.
Taking a RAID5 hard drive array with a large number of hard drives as an example, it is likely that several of the hard drives purchased at the same time have become unstable. The server’s Offline prompts may not be the most accurate, and often misreport (also due to hard drive instability) or miss other hard drives that also have hidden problems. In the seemingly normal process of forcing a hard drive online, the drive needs to perform a lot of read/write operations, which is a serious test: if you’re lucky, you’ll get through it, but if you’re unlucky, you’ll cause the other hard drives to suddenly drop off the line, which makes the problem even more difficult.
Of course, when a drive drops out, or drops out within the RAID’s fault tolerance, forcing it to come online is still a must. But before doing so, it’s highly recommended to check the stability of each drive to be on the safe side. Don’t overlook such a small detail, often this step can save the RAID server from paralysis.
2. The key to data security: do nothing?
For manufacturers, all warranty commitments are limited to hardware equipment, that is, hard disk array failure caused by the data security crisis has nothing to do with it, is responsible for the user to replace or repair has been considered “due diligence”. However, what users need is not to retrieve the hard disk or rebuild the RAID, but the “unrepeatable” data. If you look for the machine manufacturers, this is like “asking for trouble”, because the so-called after-sales service is not from the user’s original needs, as soon as possible to ensure that the hardware equipment is intact, but at the expense of the data. Of course, well-known brands of server manufacturers or agents will take the initiative to choose for large customers RAID array data recovery service fee “bill”, but this is only an individual case.
According to the introduction, because of the service provider’s after-sales service engineers to find a repair and lead to the final data completely destroyed the case almost every week. After all, after-sales service engineers are basically biased in favor of the hardware itself, and in the face of software-based operation of the RAID array data recovery is also slightly insufficient.
In addition, even a simple test operation may bring inconvenience to the later array data recovery. In the case of an ordinary RAID5 hard drive array data recovery, for example, when the hard drive is damaged beyond fault tolerance, the damaged hard drive’s image must be completely extracted for the final array data recovery. And if the hard disk is powered up frequently, it is easy to cause further fatigue damage to the magnetic head or even scratch the platters. When faced with this kind of data security problem, the right way to deal with it should be to consult a professional data recovery service provider immediately.
3, hard disk array data recovery shortcut: seek professional data recovery service providers to help
According to relevant information, many cases of data recovery failure is because the user has taken an incorrect solution. For example, some server after-sales service, they deal with this kind of array data recovery cases, easily choose to rebuild, resulting in more serious data loss, which is the enterprise data security managers should pay attention to in the face of major server array data recovery.
In addition, when choosing a data recovery service provider, judging based on size as well as industry reputation is a viable shortcut. For example, efficiency source (international) technology, franchisee service outlets throughout the country, has a deep accumulation, can help encounter RAID array data disaster users in time to get through the difficult times, professional data recovery personnel for server hard disk array data recovery.
Qunhui ds415+, 4 hard disks, raid5 recovery, please god help to recover, the data is a bit useful, thanks. The solution can be red packet
There may be a variety of reasons for the failure of RAID5, or RAID controller failure, or a sudden power failure caused by the RAID information error, it is also possible that one of the hard disk of the RAID5 error, did not replace it in time, and waited until the second hard disk error, resulting in the failure of the RAID5. The first case, RAID5 hardware failure, then this article can not help, but the latter two cases, as long as the mastery of the method, the operation is appropriate, the data can still be retrieved.1, first SCSI hard disk enclosure directly connected to the SCSI expansion card does not contain a RAID function.
2, and then engineers in a dedicated (windows2003 after modification) repair platform to connect all the hard disk in the server array in a single disk.
3, in order to protect the customer’s original data, to avoid data recovery operations to change the customer’s original data, in all data recovery process must be all hard disk in a read-only way to do a complete mirror, while the mirror is also stored in the device with redundancy protection.
4. Arrange for engineers to analyze the double-cycle checksum parameters of the original RAID from the mirror file to build a virtual raid platform.
5, in the virtual RAID platform to remove the early offline disk, explain the file system, at this time you can already export raid data.
6, in the customer’s original HP server connected to the disk array, reconfigure the RAID.
7, and finally through the network dd, NFS, SAMBA, FTP, SSH and other data transfer methods to all the data back to the newly built RAID array. 100% of all data in the Raid recovery success, all work lasted 2 days.
Double the speed of your hard drive and make it safer: What is Raid?
Double the speed of your hard drive and make it safer: What is Raid?
Liu Zongyuan
No. 19021210731
[Nested Nose] Let’s talk about what Raid is all about, what it can do, what are the advantages and disadvantages, and how to implement it.
[embedded cattle nose] Raid, hard disk
[embedded cattle question] Raid is what
[embedded cattle body] What is Raid
We first come to realize the concept, Raid full name is RendantArraysofIndependentDrives, the Chinese name for the disk array, directly translated as “redundant array of independent drives. Simply put, it is a logical array of multiple hard disks. Raid is the combination of multiple independent hard disk rows into a single logical array, as a whole to use to achieve improved transfer speeds, security and other features, mostly used in file servers or NAS.
Types of Raid
Below we introduce the more common Raid0/1/10/5/6 and the more advanced Raidz/Higher Raid, in turn. more advanced Raidz/z2, and describe how they work and their advantages and disadvantages. Uncommon Raid2/3/4 and a variety of strange combinations such as Raid100/30/50/60, etc. will not say more, understand the basic, the combination of natural understanding. If not labeled, the default is the same capacity for different hard drives.
1, Raid0
Raid0 requires a minimum of two.
Taking two hard disks as an example, the capacity of the group after the Raid is the sum of the two, when reading and writing data, read and write to both hard disks at the same time.
The advantage is that the bandwidth is doubled, and theoretically the read and write speeds are also doubled. However, the disadvantage is also fatal, because the overall data is written on the two hard disks at the same time, and read at the same time to get the complete data, so as long as one hard disk in the array is damaged, the entire array of so all the data is lost.
In summary, speed x N, risk x N, 100% utilization.
2. Raid1
Raid1 requires at least two hard disks.
The same two hard disks, for example, have a capacity equal to the smallest one, and the same for multiple ones. When writing data, the same data is written to all the hard disks at the same time, which is equivalent to mirroring or backing up, and if one of the hard disks is corrupted when reading, the complete data can be read from the other hard disks in the array.
The advantage is double the redundancy and security, the disadvantage is low utilization.
Summary, speed x 1, risk x 1/N, utilization 1/N.
3, Raid10
As the name implies, it is Raid1+0, a combination of Raid1 and Raid0, so that the two complement each other. A minimum of four hard disks are required, or a flexible combination of m×n blocks is fine, precisely because it is a combination of two Raid modes.
Take the example of four hard disks, two for backup and two for add-on.
First of all, I would like to mention the special implementation of Raid10, which is called Raid10 instead of Raid01 for a reason. You need to divide the four hard disks into two groups, and within the groups, you can first perform a combination similar to Raid1, which is a mirror of each other and is called “mirroring”, and then you can use it for two “Raid1 arrays”, and then you can use it for two “Raid1 arrays”. “Raid1 array” similar to the combination of Raid0, known as “do stripes”. So when reading and writing, read and write to four hard disks at the same time, where each group writes different data (data in different locations in the original data, may be the same, such as the first two 1s in the 1101, the same below), and two hard disks in each group writes the same data.
The advantage is to take into account the transfer speed and security, the disadvantage is that the performance overhead is large, and …… amount …… expensive, right as the disadvantages of the program.
Summary, with four hard drives, for example, speed x 2, risk x ½, utilization rate of 50%.
Similarly, the principle of Raid01 can be understood in the opposite way, first do the striping, then the mirroring.
4, Raid5
Raid5 requires a minimum of three hard disks, which need to come up with the equivalent of a hard disk capacity to store the checksum code, the checksum code using the parity mode of calculation, the checksum code can not be all in a hard disk, the rest of the data used to store. Used to store the check digit capacity from all the hard disk from different locations to take out the same capacity, add up to equal the capacity of a hard disk to three hard disks, for example, three hard disks to take out ⅓ of the capacity of each of the three hard disks, was taken out of the three hard disks are not all in the same location.
When writing data, write different data on two hard disks at the same time, and the last hard disk writes the calculated check digit, the check digit of the next write must not be stored in the same hard disk with the check digit of the previous write. If a hard disk is damaged, only the damaged hard disk needs to be replaced, and by evenly distributing the data and checksum in each hard disk, the data in the damaged hard disk can be recovered and rebuilt.
The advantage is a certain degree of balance between transfer speed and security. The disadvantage is that in reality the data recovery speed and success rate is not very satisfactory. There are also limitations that have to be mentioned, no matter how many hard disks are in the array, damaging two at the same time is unrecoverable, and all the data in the entire array is lost.
Summary, speed x (N-1), the risk will not be calculated, some day to ask the science college seniors …… can be sure that the risk is greater than the Raid1, the utilization of (N-1)/N.
Additionally, if the hard disk capacity of the different, Raid5 will take the capacity of the smallest capacity hard disk in each drive and use the rest, but Qunari’s SHR technology can utilize the wasted capacity, for those interested.
5. Raid6
Raid6 can be said to be the upgraded version of Raid5, which requires a minimum of four hard disks, two of which are used for storing the two parity codes, which are assigned in a similar way to Raid5.
Similarly, damage to three hard disks at the same time makes the entire array unrecoverable.
Summary, speed x (N-2), the risk will not count but smaller than Raid5 than Raid1, the utilization rate of (N-2)/N.
6. Raidz/Raidz2
The two “Raid mode
Raidz/z2 is based on the more advanced ZFS filesystem, which has a more advanced redundancy mechanism that solves the Raid5/6 “full rewrite” problem. This is not a moment and a half can not finish, for the sake of space and do not overpower the main consideration, for the time being, leave a pit.
Implementation
The implementation of Raid arrays is divided into hard and soft implementations, and the arrays composed of them are known as hard Raid and soft Raid.
Let’s take a look at each:
1, hard implementation/hard Raid
. p>Hard implementation is the way to realize the Raid array through hardware (Raid card). Reading and writing data to and from the disks is done by the Raid card master, and the system recognizes the entire array as a single hard drive.
The advantage is that it relies very little on the CPU and has better performance, and a cached Raid card can also improve random read and write performance.
The disadvantage is that it is not flexible enough, after using a Raid card to set up a Raid, the hard disks in the array can not be used if they are not unmounted/unmounted/canceled and mounted directly on another computer. As well as the need for additional hardware, the cost is higher.
2, soft realization/soft Raid
Soft Raid is Raid realized by software, including setting from BIOS. The reading and writing of disk data is done by the CPU, and the system can recognize all the hard disks in the array.
The advantage is that it is flexible in use, and can be removed and used directly on other computers at a lower cost and easy to operate.
The disadvantage is that it is more CPU-dependent and has poorer performance.
Understanding the above, partners who want to group Raid should have a bottom in their hearts, and can choose the way they want on demand. Lastly, IT Home would like to remind one more thing, data ten million, safety first. Raid is not standardized, their own two lines of tears.