Remediation of web site inaccessible after hard drive damage

Access | Hard drive Remember when I was in www.itnow.com.cn, the site of the server often appear a little problem, sometimes the speed is very slow, sometimes OA can not, sometimes the site can not visit, encounter bad luck, OA and ITNOW.COM.CN are not login, this not only affected the normal operation of the site, but also reduced the access to the site Amount, plus Itnow server some space to rent to other sites, those sites need to run 24 hours a day, when the server can not run normally, also affected their work, so they often complained. There have been several times, the server in the middle of the night, the results of the technical staff need to get up immediately to repair, if a big problem, technicians tried to sleep all night.

Later this matter to the technical staff, the original assembly server, SCSI hard disk 3 years after the use of aging, intermittent off-line, resulting in inaccessible.

A typical server will do RAID 1 or RAID 5 backups, and if one of the hard drives in RAID 5 is offline, it will still be accessible, but the access speed is significantly slower due to the need for other hard drives to compute the data that is damaging the hard drive.

The slow speed of writing data is more obvious if you are backing up a RAID 1, the slowdown in speed is not as obvious as Radi 5.

If the damaged hard disk is the system disk, if it is in raid, the system does not crash immediately and continues to function.

However, if the system disk in the raid is also the boot disk, it will not start properly when the server is restarted, and a "blue screen" will appear.

In general, the speed will be reduced, regardless of whether any of the hard drives in RAID 1 or RAID 5 are damaged, but at least the server will remain operational.

Hard disk damage is not a precaution, only if there is damage to remedy.

When an administrator discovers that server access slows down, you can check for damage to the disk in Disk Manager.

The administrator can first attempt to reactivate the hard disk, but if the disk is still frequently offline, the hard drive may have been physically damaged and must be replaced as soon as possible. Otherwise, it is difficult to salvage when a second hard drive is damaged in raid. This is the system crashes, data loss, the site can not be accessed.

Dedicated servers with a hard disk rack, the server can be hot-swappable SCSI hard drives without shutting down the computer, or you can replace the hard drive without pausing.

Or use a hardware-level raid backup, with a separate RAID card, one hard drive for reading, the other hard disk only for backup, and when the raid card finds a working hard drive problem, automatically starts the backup disk into the work disk.

The 3rd option is to use a tape drive, like a tape recorder, to write data directly to a tape backup, but to manually or by dispatching a tape drive to back up data, but this does not even solve the hard disk corruption problem.

Usually the tape drive is to solve the software-level or hardware-level RAID backup can not solve the problem, such as: misoperation deleted data, multiple hard disk damage, can not recover when used.

Recovery requires a new hard drive to be replaced, and the data is read back from the tape drive to the hard disk before the server can be rerun. It takes a long time to recover. And the tape drive is generally an external device and is not part of the server itself.

However, I asked him why sometimes the accident occurred so often, he said that the replacement of all the second-hand hard drives, may be to save money 、、、