Electrical computers and digital processing systems: multicomput – Computer network managing – Computer network monitoring
Reexamination Certificate
1998-10-01
2002-03-05
Maung, Zarni (Department: 2154)
Electrical computers and digital processing systems: multicomput
Computer network managing
Computer network monitoring
C709S220000, C714S004110
Reexamination Certificate
active
06353854
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to computer systems coupled to a network and more particularly to a system having logic to receive transmissions across a connection to a network to allow for automatically updating configuration information and system management parameters.
2. Description of Related Art
Personal computer systems are well known in the art. Personal computer systems have attained widespread use for providing computer power to many segments of modem society. A personal computer (PC) can typically be defined as a desktop, floor standing, or portable microcomputer that includes a system unit having a single central processing unit (CPU) and associated volatile and non-volatile memory, including random access memory (RAM) and basic input-output system read only memory (BIOS ROM), a system monitor, a keyboard, one or more flexible diskette drives, a CD-ROM drive, a fixed disk storage drive (also known as a “hard drive”), a “mouse” or pointing device, and an optional network interface adapter. One of the distinguishing characteristics of such a system is the use of a motherboard or system planar to electrically or operationally connect these components together. Examples of such PCS are computer systems within IBM's PC 300 series and IBM's IntelliStation Series. The PC of the before referenced related patent document Serial No. 104835 identified in the illustration of
FIG. 3
thereof and as described in the patent document specification thereof, is an example of a typical client computer system.
With PCS being increasingly connected into networks to allow transfers of data among computers, more operations such as maintenance, updating of applications and data collections are occurring over the network. As computers are also becoming more and more essential to their users, it is desirable to minimize loss of productivity by increasing the availability of PCS. This includes detection and reporting of intermittent failures on a system that will allow system administrators to schedule maintenance for the PC at a convenient time. In addition, the immediate detection and reporting of an inoperable PC is required, since it has an immediate impact to productivity. There is no reason to wait until employees arrive on the next working day to discover that the machine failed yesterday or over the weekend.
One solution to this problem has been alert technology, such as IBM's Alert on LAN technology, which supports detection and reporting of failures over a network. Alert on LAN solution creates network alerts to provide event and status information to a network administrator. The technology detects and reports events such as operating system (OS) hang, POST/BIOS error codes, and voltage and temperature problems.
When the client systems are deployed, the network administrator must configure the alerts supported and the TCP/IP Internet Protocol (IP) information for the administrator's management console. This information is then stored in a non-volatile EEPROM in the client system.
It is normal to custom configure the system unit of a computer system for a user at a corporate configuration center or at the user site. When done at a configuration center the resources and skills needed are available, but the computer must be removed from the packaging, set up, configured with BIOS and a program image, and configured to support system management applications such as Alert on LAN and then repackaged to be sent to the end user. At the time of configuration, the final location or owner of the system may not be known. Therefore at the time of configuration the IP address of management server is not possible, given that most networks have several management consoles. The missing critical pieces of information will require re-configuration at the final location.
At the user site, roll out and updating of systems is a less efficient use of resources but avoids the determining-final-location problem. For complex system management program images, it may be necessary to send out a skilled configuration expert thus consuming time when the expert's valuable skills could otherwise be put to use, as the expert travels to the user site.
Accordingly, a number of methods have been developed for simplifying the remote configuration of computer systems. For example, the Desktop Management Interface (DMI) industry standard provides a mechanism to retrieve and update remotely from a client personal computer. The DMI technology is widely used in standard system management applications such as Intel's LANDesk programs and IBM's NetFinity program.
The task of re-configuring IP headers using DMI is accomplished by the following steps. First the management server must discover all the systems on the network. The management server can only do first time discovery of systems on the network if they are in full power on state running OS with a management application with DMI capabilities such as Intel's LANDesk Client Manger or IBM NetFinity. After the system is discovered it is placed into a database. Next the systems must be placed in a full-on power state, since DMI runs on the OS. This may be accomplished using Wake on LAN technology. Finally DMI is used to re-configure the IP header by sending commands to the Alert on LAN management application. This is a complicated and time consuming operation, given that several hundred machines are set up to send alerts to a single server. This mechanism also requires the system administrator to gain control of each system to accomplish the update.
The time required to update each system unfortunately provides an opportunity for loss of information, since information sent by a client, prior to being reconfigured to the new server, is lost. There is a potential significant down-time when there is a change in server due to a crash or network issue. Each client system must be reconfigured with the IP address of the new management server. In the case of routine maintenance, it is possible to avoid loss of information, but the reconfiguration process is very time consuming and potentially disruptive to operation on the client.
What is needed is a robust and effective mechanism to update the destination IP address in the client PCS to point to the new server. The present invention provides for such a mechanism and will update the IP address regardless of the power state, i.e., full-on, sleep, or off, of the client. In addition, the mechanism provides for an advantageous method for handling updates to groups of systems within a network.
SUMMARY OF THE INVENTION
The present invention recognizes that by designing improved hardware reception capabilities in the computer. It is possible to reconfigure a system unit at a remote location without requiring DMI. The invention involves the recognition that a system normally, and for good reason, protects itself from being re-configured remotely, and it is desirable to override these protections and have the computer automatically reconfigure based on a special packet from a management console.
One embodiment of the present invention is directed to a method for detecting re-configuration packets from a management console and then updating hardware configuration. The system, when necessary, is supplied with auxiliary power and is operative to receive and accept updates regardless of the system power state. The method includes providing a computer system connected to a management console via a data transmission network. The invention includes a detection scheme to recognize re-configuration network packets. The invention further includes a scheme for the computer to handle re-configuration of sub-groups of machines.
The present invention recognizes that certain normal reconfiguration operations, that involve updating operations as opposed to initialization operations, are minor operations but nonetheless are expensive and time consuming because, for example, a service person must come to the site or use a time consuming process, such as employing DM
Cromer Daryl Carvis
Ellison Brandon John
Kern Eric Richard
Locker Howard Jeffrey
Ward James Peter
Maung Zarni
Schelkopf J. Bruce
LandOfFree
Automatic reconfiguration system for change in management... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Automatic reconfiguration system for change in management..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Automatic reconfiguration system for change in management... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2831673