US20070100973A1 - System and method for reliably purging a fault server - Google Patents

Abstract

Improvements to existing trap-generated message memory purge procedures and processes are shown and described. The improvements may be implemented in a telecommunications system having a plurality of managed elements, each of the managed elements potentially generating traps which are communicated to one or more fault servers.

Description

I. BACKGROUND
• A. Field of the Invention
• This invention relates generally to the field of network management, and more particularly to maintenance operations on elements within a managed telecommunications network.
• B. Copyright Notice/Permission
• A portion of the disclosure of this patent document contains material that is subject to copyright protection. The copyright owner has no objection to the reproduction by anyone of the patent document or the patent disclosure as it appears in the Patent and Trademark Office patent file or records, but otherwise reserves all copyright rights whatsoever. The following notice applies to the software and data as described below and in the drawings hereto: Copyright. COPYRGT.2001-002, BellSouth Intellectual Property Management Corporation.
• C. Description of the Related Art
• Telecommunications companies (i.e., service providers) build, operate, and maintain very large communications and related networks. Part of the operation and maintenance of these networks involves the use of operations software, typically divided into a number of functional areas such as engineering, provisioning, and the like. Provisioning software aids service providers in receiving requests for service or alterations to existing service, be it voice and/or data, and configuring both the telecommunications network and/or related networks and systems (e.g., accounting, billing, and the like) to provide the new service requested. Engineering operations software in contrast is typically used by service providers to configure and monitor network elements to ensure they perform their functions properly. Service providers also use engineering operations software to facilitate service provisioning and monitoring.
• One of the primary engineering operations software systems is the element management system (EMS) software. Typical EMS packages are centralized service network management applications that manage and control (typically via standards such as SNMP and the like) the various elements in the telecommunications and/or related networks. Within the core telecommunications network the elements often are multiservice elements such as frame relay, SMDS, ATM, IP, and/or the like switches. Some of the operations performed by typical EMS packages include: circuit provisioning to establish end-to-end network connectivity; logical provisioning of individual circuits and to establish network-wide parameters; providing audit trails on activities such as the length of a user session and the addition or modification of switches, logical ports, trunks, circuits, and the like; display of network statistics for real-time status information on logical and physical ports; display of usage data on logical and physical ports and the like for network planning and trend analysis; and collecting different types of traps for alarm indications and statistics logging for the numerous objects in the telecommunications networks (e.g., switches, trunks, physical ports, logical ports, permanent virtual circuits, switched virtual circuits, and the like).
• With regard to traps in particular, the EMS package typically reports all traps from the various elements in the network being managed to a central repository comprised of one or more fault servers and/or related databases. However, with the explosive growth in demand for telecommunications services over the past few years the number of elements within the service providers' networks have dramatically increased. As a result, the number of faults occurring in service providers' networks has swelled, thereby generating so many traps at a such a rapid pace that existing systems and methods of collecting, analyzing, and managing these traps have been overwhelmed. Accordingly, there is a need for improved systems and methods of collecting and managing traps in telecommunications and/or related networks.
II. SUMMARY OF THE INVENTION
• In a telecommunications system having a plurality of managed elements, each of the managed elements potentially generating traps which are communicated to one or more fault servers, an improved fault message purge procedure, the improvement comprising an increased rowcount, the increased rowcount corresponding to approximately 45,000 rows in a trap-generated message memory for approximately every traps received at the one or more fault servers. The purge procedure may call a purge script residing in the one or more fault servers. The purge procedure may also be initiated by a second script residing in a UNIX segment of the one or more fault servers. Existing purge procedures are improved by monitoring one or more of any processes contained within the purge procedure and restarting the purge procedure upon detection of any errors in the processes.
III. BRIEF DESCRIPTION OF THE DRAWINGS
• These and other features, aspects, and advantages of the invention will become better understood in connection with the appended claims and the following description and drawings of various embodiments of the invention where:
• FIG. 1 illustrates and an exemplary network within which the invention may be implemented; and
• FIG. 2 illustrates the structure of an exemplary server that may reside within a network such as that illustrated inFIG. 1.
IV. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
• Throughout the following detailed description similar reference numbers refer to similar elements in all the figures of the drawings.
• FIG. 1 illustrates anexemplary network101 in which the invention may be implemented. Network101 is based in part on the EMS developed and marketed by Lucent Technologies of Murray Hill, N.J. under the trademark NAVISCORE. The NAVISCORE EMS is a distributed multiservice element manager that utilizes a graphically integrated UNIX-based platform and telecommunications network management (TNM) standards to perform its network management and control functions. Network101 also includes portions of a suite of management servers developed and marketed by Lucent Technologies under the trademark NAVISEXTEND ENVIRONMENT. The NAVISEXTEND ENVIRONMENT extends the functionality of the NAVISCORE EMS.Network101 as depicted includes a plurality offault servers102 andstatistics servers103 operatively connected to aprivate network104. Network101 also includes afault database105 and astatistics database106 operatively connected toprivate network104. As will be understood by one skilled in the art,network101 need not include many of the elements depicted therein (e.g.,statistics servers103, firewalls,DMZ network108, and the like), and may include any number of other elements not depicted inFIG. 1 (e.g., provisioning servers, accounting servers, and the like).
• In operation, whenever a switch or managed network element (not shown) in thetelecommunications network107 experiences a fault it generates a trap. The trap is subsequently communicated from the network element to at least one of thefault servers102 via a demilitarized zone (DMZ) network and theprivate network104. Thefault server102 converts the trap into an English language-type message (not shown) that typically includes information such as the type of error experienced by the network element, a date and time the error occurred, the particular network element that experienced the error (e.g., by network address such as an IP address), and the like. In some of the assignee of the present invention's networks, receipt of 50-100 traps per second at thefault servers102 is not unusual. The English language-type message is then sent by thefault server102 to thefault database105 via theprivate network104, where the message is stored and may be accessed by other systems in the network for analysis, troubleshooting, and the like.
• While one skilled in the art will understand thatservers102 may be implemented in any number configurations on any number of computing platforms,FIG. 2 illustrates ageneric computing platform201 forservers102. As shown,computing platform201 includesprocessing unit222,system memory224, andsystem bus226 that couples various system components includingsystem memory224 to theprocessing unit222. Thesystem memory224 might include read-only memory (ROM) and/or random access memory (RAM). Theplatform201 might further include a hard-drive228, which provides storage for computer readable instructions, data structures, program modules, other data, and the like. A user may enter commands and information into theplatform201 through input devices such as akeyboard240 and pointing device242. Amonitor244 or other type of display device may also be connected to theplatform201 for visual output.Communications device243, which may be for example a TCP/IP enabled device, provides for connectivity to other computing devices within or beyondnetwork101 illustrated inFIG. 1.Processor222 may be programmed with instructions to interact with other computing systems so as to perform the algorithms and operations described below.Processor222 may be loaded with any one of several computer operating systems such as Windows NT, Windows 2000, Linux, and the like. In a particular embodiment of the invention,processing unit222 comprises a 4×450 MHz CPU,system memory224 comprises 4 Gigabytes of RAM, hard-drive228 comprises a 36 Gigabyte disk-drive, andprocessor222 includes a UNIX segment.
• Because the information contained in the stored messages generated from the traps becomes stale at some point and the amount of storage space in thefault database105 is necessarily limited, a purge script is run periodically to expunge a predetermined number of older error messages stored in thefault database105. In one configuration of thefault servers102 the purge script calls on a Sybase stored procedure that resides in a UNIX-based segment offault database105. Optimally, older error messages would be kept for the duration of their usefulness while no fresh error messages would be lost due to insufficient storage space in thefault database105. The developers of existing purge scripts however failed to anticipate the sheer number of traps likely generated by the elements in service providers' networks. The existing purge scripts therefore failed to allocate enough system resources to handle the volume of traps generated in current networks, failed to purge an adequate number of stale messages stored in the fault servers, and/or failed to provide for the appropriate periodicity of execution.
• We have determined a number of ways that existing purge scripts may be improved so that a more appropriate number of stale or older stored messages are expunged, a more appropriate number of newly generated messages from traps are retained in memory, and the periodicity of the purge process is adjusted to ensure no system errors are generated because insufficient system resources are available to the purge process and/or the process is overwhelmed by the sheer number of messages being generated in response to traps received from the various networks. Typically memory within a database or memory table is allocated by row. We have determined that in a database or memory where a row comprises approximately 1 kilobytes of memory for alarms and about 1.5 kilobytes of memory for traps (generated from alarms), and there is approximately 5 Gigabytes of memory allocated for storage of up to ten days worth of traps and alarms, purging the last 45,000 rows of memory will free adequate storage space where a fault server(s) receives approximately 15 traps per second from the various networks reporting to it, and where the purge process or script is mm approximately hourly. For example, in one embodiment of the invention where thefault servers102 are receiving approximately 50-100 traps per second, the purge script is run hourly with a rowcount set to free or return up to 1,500,000 rows of memory infault database105. Pseudocode for a revised purge script (“fs_purge.script”) appears in Appendix A attached hereto. In an exemplary embodiment of the invention a Unix script (“fsPurge.sh”) residing in a UNIX segment offault servers102 is the procedure that calls or initiates the purge script (“fs_purge.script”) which resides in thefault database105. Pseudocode for exemplary “fsPurge.sh” instructions is attached hereto as Appendix G.
• Another improvement we have determined can be made to existing purge procedures is the addition of instructions to the procedure or process that initiates the purge script. Some of these additional instructions count each insertion and deletion of a trap-generated message from memory in hourly periods and then place the data gathered in a log file (“fs_inserts.script”, “fs_stats.script”, and “fs_stats_hr.script”). This insertion and deletion data subsequently may be analyzed for troubleshooting or optimization of the purge process. Pseudocode for exemplary embodiments of these additional instructions appear in Appendices B, C, and D attached hereto.
• Another set of additional instructions that may be added to the purge procedure is a script that monitors the fault server processes related to purging operations and automatically restarts them if problems are detected such as a fault database deadlock message. Pseudocode for exemplary embodiments of these additional instructions (“fault_cron” and “check_insert.sh”) appear in Appendices E and F attached hereto. Note that these two scripts monitor the log file noted above in conjunction with the fs_inserts and fs_stats scripts.
• Note that the exemplary embodiments of the invention illustrated in the various appendices attached hereto are designed for the purge procedure to be run hourly, preferably every hour on the hour. Note also that instructions for the exemplary embodiments depicted in the appendices also provide for the purge procedure to restart up to ten times, separated by one minute intervals, in the case of fatal errors. This helps to ensure that a complete purge is completed even if the purge script and/or the procedure it calls deadlocks or is killed by the server or database respectively.
• While the invention has been described in connection with various exemplary embodiments depicted in the various figures and appendices, it is to be understood that other embodiments may be used or modifications and additions may be made to the described embodiments for performing the same function of the invention without deviating therefrom. The invention therefore should not be limited to any single embodiment, whether depicted herein or not. Rather, the invention should be accorded the full breadth and scope encompassed by the claims appended below.

Claims (13)

1. A fault message purge procedure comprising the improvement of an increased rowcount, the increased rowcount comprising approximately 45,000 rows in a trap-generated message memory for approximately every 15 traps received at a corresponding fault server.

2. The procedure ofclaim 1 comprising the further improvement of maintaining a log file of all insertions and deletions of trap-generated messages in the trap-generated message memory.

3. The procedure ofclaim 1 comprising the further improvement of monitoring all insertions and deletions of trap-generated messages in the trap-generated message memory to ensure all purge processes are functioning properly.

4. The procedure ofclaim 4 wherein any of the purge processes identified as not functioning properly are restarted.

5. The procedure ofclaim 1 comprising the further improvement of monitoring one or more of any processes contained within the purge procedure and restarting the purge procedure upon detection of any errors in the processes.

6. A fault message purge procedure comprising the improvement of monitoring one or more of any processes contained within the purge procedure and restarting the purge procedure upon detection of any errors in the processes.

7. In a telecommunications system having a plurality of managed elements, each of the managed elements potentially generating traps which are communicated to one or more fault servers, an improved fault message purge procedure, the improvement comprising an increased rowcount, the increased rowcount corresponding to approximately 45,000 rows in a trap-generated message memory for approximately every 15 traps received at the one or more fault servers.

8. The procedure ofclaim 7 wherein the purge procedure calls a purge script residing in the one or more fault servers.

9. The procedure ofclaim 7 wherein the purge procedure is initiated by a second purge script residing in a UNIX segment of the one or more fault servers.

10. The procedure ofclaim 7 comprising the further improvement of maintaining a log file of all insertions and deletions of trap-generated messages in the trap-generated message memory.

11. The procedure ofclaim 7 comprising the further improvement of monitoring all insertions and deletions of trap-generated messages in the trap-generated message memory to ensure all purge processes are functioning properly.

12. The procedure ofclaim 11 wherein any of the purge processes identified as not functioning properly are restarted.

13. The procedure ofclaim 7 comprising the further improvement of monitoring one or more of any processes contained within the purge procedure and restarting the purge procedure upon detection of any errors in the processes.

Images