Shulou(Shulou.com)06/02 Report--

Raid disk arrays are listed as servers that provide secure, reliable, and scalable external storage space. However, most server users do not know much about raid, and the advertisements of the server put too much emphasis on the mistolerance function of the raid array, so that many server users are misguided by a kind of "raid will not fail". Therefore, the potential risk of raid disk array is ignored in the use of the server, and the backup or emergency plan of the server data is insufficient. Once the server fails, it will have a great impact on the enterprise. The reasons for the failure of the Raid disk array can usually be attributed to the following:

Raid controller failure leads to raid failure, sudden power outage leads to raid information failure, Raid5 one hard disk error, the system administrator does not change the hard disk in time, another hard disk error leads to raid5 failure.

The following will introduce readers to the server disk array in the form of raid1, raid0, raid5 three forms of raid failure of data recovery ideas and methods.

1.raid1 array data recovery

Raid1 is the simplest form of all raid arrays. Two hard drives in raid1 mirror each other, and all the data are exactly the same. If the data of the raid disk array cannot be accessed because of the failure of the raid1 controller or the error of the raid information, you only need to remove one of the two hard drives from the server and mount them to the computer as a separate hard disk to read the data.

If one of the hard drives in the raid1 fails, it will not affect the normal operation of the server, then you need to replace the failed hard disk with a normal hard disk. If one hard drive of the server fails to replace the hard disk in time and the other hard drive fails, the raid1 disk array fails. At this time, if you want to recover the data, you need to use the hard disk that failed to recover the data.

The idea of 2.raid0 disk Array data recovery

Raid0 is the most vulnerable form of disk array in all disk arrays, raid0 disk array does not have any redundant performance, as long as there is a hard disk failure in the array, the server data will be lost, so it is a very risky form of array. As can be seen from the following figure, the data of raid0 is distributed to each disk, and if any hard disk in the server fails, the data of the server will be incomplete.

The data recovery of raid0 disk array requires data recovery engineers to reorganize all the data in the array, and because the raid0 array is no longer available, the hard disk can only be taken out of the raid controller as a single hard disk for analysis and data recovery.

As shown in the figure above, for a single hard disk 1, the data in hard disk 1 is A/E/I/M, and the data in hard disk 2 is B/F/J/N, which is only part of the data. Only when all the hard disk data in the array are spliced in such order as A/B/C/D/E/F/G/ H, can we really restore the data in the raid0 array.

So how to splice all the hard disk data in the raid0 array sequentially? here we need to pay attention to two factors. One is the size of each data block in the raid0 disk array, that is, the number of sectors occupied by An or B data blocks. Another factor is the order of the hard drives in the raid0 array, that is, you need to determine which hard disk is the first hard disk in the array, which is the second or third hard disk.

Taking the raid0 disk array in the above figure as an example, we assume that the size of the data block is 16 sectors, and the order of the hard disk takes the one in the figure as an example, then we only need to extract the information of 015 sectors in hard disk 1, then extract the information of 015 sectors in hard disk 2, then extract the information of 15 sectors in hard disk 3, and then extract the information of 15 sectors in hard disk 4. Then go back to hard disk 1 to extract the information of sector 16'31, and so on, all the data in this raid0 array can be extracted.

Principle of 3.raid5 disk Array data recovery

The distribution of data in the Raid5 array is similar to that in the raid0 array, except that one of every parallel data block in the raid5 is always a parity block, such as p1/p2/p3/p4 in the following figure. Raid5 supports normal data access when one hard disk is offline, but if two or more hard disks are offline at the same time, the array will fail and the disk array needs to be reorganized. The data reorganization of Raid5 is the same as that of raid0, only the data in the hard disk need to be spliced in order.

Because every hard disk in the raid5 array has check information, it is necessary to analyze the position and direction of one more check block than the raid0 array when analyzing the raid5 array. In other words, there are three factors in raid5 array analysis, namely, the order of hard drives, the number of sectors occupied by each data block, and the size of each data block in the array.

Taking the raid5 array in the above figure as an example, assuming that the data block size of the raid5 array is 32 sectors, and the disk sequence is as shown in the figure, then when extracting the data, you only need to extract the information of the 0mm 31 sector from 1 to 4 respectively (the information of the 0ft 31 sector in the hard disk 4 is the check block, skip it), and then return to the hard disk 1 to extract the 32 sectors 63 sector information, and so on. All the extracted data can be formed into a complete raid5 array image file.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.