одна из них работает, вторая не работает но включена, на случай аварии
основной.

в памяти (в 98 - 99 пробовал на winNT) осталось, что виндюки стартуют на второй машине сами, при выключении первой
:(((

cyc10ne а покажи пальцем если не лень где почитать

в двух машинах - один винчестер
машины включены в сеть и работает одна из них.
если она выключается - вторая автоматически стартует и замещает первую.

какой какой сказик. шлейф одним концов в одну тачку другим концом в другую в середине винт


Basic Cluster Configuration
The nodes in an MSCS cluster are connected using one or more shared storage buses and one or more physically independent networks (sometimes referred to as interconnects). If one network does not connect clients to the cluster, but serves only to provide a redundant path for intracluster communication, it is referred to as a private network. In this case, the network that supports clients is referred to as the public network.

Note Although MSCS clusters can function with only one interconnect, two interconnects are strongly recommended to eliminate a single point of failure, and are required for the verification of MSCS original-equipment-manufacturer (OEM) systems that include both hardware and MSCS software.

There are one or more local disks for each node. Each shared storage bus attaches one or more disks, which hold data used either by server applications running on the cluster or by applications for managing the cluster. MSCS currently supports only SCSI shared-storage buses. Each disk on the shared SCSI bus is owned by only one node of the cluster. The ownership of disks moves from one node to another when the disk group fails over or moves to the other node.

Figure 1.3 shows a typical configuration for a two-node cluster that is accessible by multiple clients on a local area network (LAN). The two cluster nodes are physically connected to the network. The cluster appears as a single network entity. The administrator tool can be MSCS Cluster Administrator (running locally or remotely) or an alternate administrative application (written using the Microsoft Platform SDK).



MSCS Failover
Failover is the process of having cluster resources migrate from an unavailable node to an available node. A related process, failback, occurs when resources are restored to the node that has been offline after it is back online. Failover is automatically initiated by the Cluster Service when it detects a failure on one of the cluster nodes. Because each cluster node monitors both its own processes and the other node, the need for failover is detected without delay. In most cases, MSCS can detect a node has failed and begin failover in less than 10 seconds.

Figure 1.4 shows the failover process. A node on the right has failed, causing FilePrint group2 to fail over to the node on the left.

я пробовал на NT4 интерпрайз эдишен для 300 целерона

юниксу такая муть наверно вообще не нужна для повышения надежности.
гы))))))))))))))))
но начальство, начитавшись инструкций по винде, интересовалось

Все ушли на NAS/SAN.

killed
куда куда?
типа на фронт?