First, alarm! Disaster is Fall ...
Bad things always happen when people least want them to appear. This day, that 4400 of the server, so suddenly died ...
According to colleagues at the scene, the server suddenly slowed down and stopped the network service until all the responses were stopped.
In order to restore the service, they only have to restart the server. After forcing the server to reboot, a blue screen appears, the System reports stop:0x0000007b error, Inaccessable_boot_device. At first I thought the MBR was broken or NTLDR and Ntdetect.com was broken, so I just need to use fixmbr and fixboot in the Recovery Console to fix it. Because the server is using hardware RAID, you can only use F6 to load RAID startup when the CD is started, and then go to the Recovery Console to see if you can access the system partition. In the big surprise, with DiskPart you can see the system partition, but the system partition is no longer readable, the dir command cannot enumerate any files and directories.
Speaking of which, I would like to introduce the server environment. This Windows Advanced server is the first domain Controller in this field (named after domian.com), with five operations master roles (FSMO) on it. Limited to conditions, Exchange Server 2000 is running above, and is also used as a file Server/printer server. In addition, there are dhcp/wins/dns. Backup tapes, insufficient capacity to complete full backups ... So, the server's exit service, for us is a real disaster. If there are no other domain Controller, we may be busy for a long time. Because there are other domain Controller, we still have a glimmer of hope!Second, Mission impossible? Impossible is nothing!0, replacing network services with other servers
Customers must be affected to the minimum. Because the DHCP server is used in the network to assign addresses to clients and provide network information, we must first resolve the issue of client DHCP lease expiration updates.
Originally we had only this DHCP server, and now it's breaking down. So we installed DHCP on another server, and the range of addresses was 20%. The reason for this is that the client that is due to renew the lease is not a lot, of course, depending on the length of the lease and the number of clients themselves, the faster the update, the more addresses are divided.