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

Preface

Recently, when I replaced the server hard disk for a user, I found that, first, the RAID configuration was wrong, and the hard disk of 300GB was used as a hot backup for the 600GB disk, and the data was not synchronized due to the incorrect hard disk model. If the newly hired administrator takes over an old server, he should focus on checking the hard disk and RAID information when he has time to see if there is any offline hard disk or improper configuration to correct it in time.

Text

In early April this year, there was a unit with 2 server hard drives that broke 3 hard drives. Customers asked to send 3 600GB hard drives and help replace them. When I arrived at the scene, I saw the two servers, one of which had two hard drives with yellow lights (shown in figure 1), and the other server had one hard drives with yellow lights (as shown in figure 2).

Figure 1. 2 hard drives turn on yellow light

Figure 2 1 hard disk with yellow light

Take a closer look, each of the two servers is equipped with 1 300GB hard disk and 3 600GB hard drives. At that time, it was quite strange how there was such a configuration. Is this piece of 300GB used to install the system and the other 3 pieces of 600GB to do RAID-5 to store data?

At that time, I told the customer that the two broken hard drives in figure 1 may need to do data recovery, because two of the three hard drives are broken at the same time, so it is not possible to change the disk at this time. At that time, the server was no longer able to start (the server in figure 2 is still in use and can enter the system).

For the server with 2 broken hard drives in figure 1, turn on the power to enter the RAID interface and see that the hard drives in 01 and 02 slots are in Failed state, as shown in figure 3.

Figure 3 two hard drives failed

After pressing Ctrl + N to enter PD Mgmt, the whole person was a little dizzy, and the 300GB hard disk of slot 0 was configured as a hot spare disk! As shown in figure 4.

Figure 4 300GB's hard disk is configured as a global hot standby

The customer said that the server had been in use for many years, and it had broken down (a hard drive). After it was replaced, it continued to be used. This time, it was only when the system could not be used that it was found that two hard drives were broken, thinking that it could continue to be used if it was replaced. At that time, the server was bought with a 300GB hard drive. People from the financial software company said that one hard disk was not safe, and they bought 3 600GB hard drives to make RAID-5, which was configured by people from the software company. It is estimated that the people in this software company did not know much about hardware at that time, so they were configured in this way.

The other server only broke one hard disk, and after replacing it, the data began to synchronize. Because the RAID configuration of the first server I saw was inappropriate, I restarted the server to enter the raid card configuration module, and sure enough, this 300GB hard disk was also configured as a hot spare disk, as shown in figure 5. At the same time, the newly replaced hard drive has begun to reorganize the data.

Figure 5 data synchronization and hot spare disk configuration

This is a photo of this server after replacing its hard drive, as shown in figure 6.

Figure 6 after the second server replaces the hard disk

At this time, the customer said, there is also a server, a year ago replaced a hard drive, this time to help check, this inspection will find a new problem.

The third server is also 3650 M5. In another computer room, the hard disk of slot 3 was replaced a year ago. This computer room is obviously better, and there is less dust on the server hard drive, as shown in figure 7.

Fig. 7 the hard drive of slot 3 was replaced years ago

Reboot into the raid card, 300GB's hard disk is still equipped with a hot spare disk, but also familiar with the formula, familiar taste. However, the hard drive of slot 3 is Unconfigured Good, and this hard drive is not synchronized, as shown in figure 8.

Figure 8 the hard drive is not synchronized

Take a closer look at figure 7 and find that the serial number on the newly replaced hard disk carrier is different from that of the original carrier (the new serial number in figure 6 is the same as the original). Fortunately, there are still two newly purchased hard drives that are not in use for the time being. Take off the hard drives purchased before the year and replace them with the ones brought this time. At this time, the hard drives begin to reorganize data, and the newly replaced hard drives flicker alternately with yellow lights and green lights, as shown in figure 9.

Figure 9 hard disk data synchronization

The data begins to be reorganized in the RAID interface, as shown in figure 10.

Figure 10 data reorganization

More than an hour later, the data reorganization is complete and the hard drive flashes green, and the server check is complete, as shown in figure 11.

Later, it is suggested that if customers need hot spare disks, each one should remove the original 300GB hard disk, add a 600GB hard disk and reconfigure it as a hot spare disk.

Figure 11 the hard drive is in good condition

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.