Operating system role
The operating system is the ultimate authority in resolving hardware problems associated with the change of centuries. It is the operating system in most cases that tells applications the current…

Continue reading →

“Clicks of Death" - what is it?
This is such a situation when the drive has lost its ability to read the servo track for proper head positioning. At the same time, the drive does not “see”…

Continue reading →

Ways of spreading viruses
How are viruses and, in particular, the Win95.CIH virus spread? It is generally believed that the main path is the Internet. Other viruses cause a lot of controversy about the…

Continue reading →

How to avoid network downtime?

A network of any company or organization is created to provide users with easy access to information, applications and network services. However, both company’s productivity and revenues can suffer significantly when a server running one or more business applications becomes unavailable. Losses from downtime, according to various sources, for an average company range from units to tens of thousands of dollars per minute. Solutions to reduce downtime, we will consider for Novell’s NetWare networks, since these networks are the most common in Ukraine.

Fault Tolerant Systems
Fault tolerance is usually understood as the ability of a computing system to perform the functions assigned to it in the event of failure or failure of some of its components. Means of achieving fault tolerance – hardware and software redundancy. It is known that to ensure the fault tolerance of servers, they use multiprocessor cards, memory with correction of single errors, hard disks combined into RAID arrays, redundant network cards, duplication of power supplies, redundancy of ventilation, means of temperature control and other vital parameters. Modern server solutions provide for a “hot swap” of most redundant components when one of them fails. The design of the buildings provides a modular architecture that helps reduce repair time. All these technological tricks are ultimately aimed at reducing network downtime.

The well-known Novell solution, in which hardware redundancy is created not only at the level of redundancy of server components, but also at the level of the servers themselves, called System Fault Tolerance level 3 or SFT III, has a higher degree of fault tolerance.

System Fault Tolerance level 3 or SFT III
In such a system, two identical servers are provided, the synchronization of which is ensured by special software and hardware.

One of the servers is primary and the other is secondary. On a network, both servers look like a single unit. In case of failure of the primary server, its mirror copy – the secondary server picks up all running processes almost instantly, so that users do not notice what happened. The secondary server in such a system operating in the normal mode does not directly participate in servicing network requests and does not increase network performance. This, as well as stringent requirements for additional hardware and software (software) subsystems that provide mirrored servers, make this solution quite expensive. This solution pays off when increased demands are placed on server availability. Requirements for complete server identity limit the ability to implement the SFT III system in practice. If you have a server that has worked for two years, then it will not be easy to find exactly the same on the market. In cases where instant switching to a mirror server is not required, it is advisable to use another solution for NetWare networks – Novell Standby Server ™.

Novell Standby Server ™
Novell Standby Server ™ is a software solution for mirroring data between servers and creating a fully redundant system to protect the network from hardware and software failures. Developed by Vinca Corporation, Novell Standby Server ™ for NetWare / intraNetWare enhances Novell’s ability to provide customers with a wide range of choices when deploying high-availability network solutions.

By mirroring the servers, Standby Server provides system redundancy and allows you to switch to the backup server at any time in the event of a hardware or software failure on the primary server. This allows you to reduce network downtime and ensure continuous user experience with business-critical applications.

Novell Standby Server ™ makes it possible to use any combination of hardware certified for working with NetWare for a mirrored pair of servers (Fig. 2).

Information is mirrored over a dedicated connection between servers using two additional network cards. Maintaining the identity of these servers is not based on copying files, but on mirroring transactions.

When the primary server fails, users who have installed 32-bit clients from Novell or Microsoft experience a slight pause while the Standby Server activation process is in progress. Users with older clients installed must re-register on the server with the same name and password.

Standby Server can be configured in One-to-Many mode, when the primary server can be mirrored to the local and remote server at the same time. The remote server connects to the primary server using WAN technology.

Golden Disks and IDE
The problem of recording "golden" discs has already been covered on our website (as well as in the journal CHIP, No. 8, 1999). However, the problems continue. Only, if earlier…

...

The Iron Solution to the 2000 Iron Problem
Before the onset of 2000, about three more months. But the situation was noticeably heated. It is understandable. If even a small fraction of what is written in the press…

...

Zip Drive Frequently Asked Questions
What are the models of Zip drives? Currently, there are 2 main types of Zip drives: Zip-100 (100 MB removable disk capacity) and Zip-250 (250 MB removable disk capacity). Zip-100…

...

Golden Disks and IDE
The problem of recording "golden" discs has already been covered on our website (as well as in the journal CHIP, No. 8, 1999). However, the problems continue. Only, if earlier…

...