Ten years ago, a product was introduced that let remote S/3X system managers know whether a system backup ran successfully. If something went wrong, the product sent a message to a pager and displayed a message indicating the trouble. From that simple beginning, an entire class of AS/400 system monitoring products has evolved that can now tell you not only the status of running jobs, but the state of your AS/400 itself.

System monitoring is the process of maintaining continuous oversight of events occurring within an AS/400 and its peripherals. System monitoring covers everything from how a specific program ran to pinpointing a specific type of hardware failure. Currently available system monitoring products can oversee multiple queues, take predefined actions such as activating commands and programs, start new jobs, maintain histories of system activity, and send messages to operators and managers via a wide variety of communications devices. NEWS/400 surveyed 10 vendors of system monitoring products and summarizes their products’ key features in the accompanying table.

System monitoring product vendors have taken two different approaches to providing the many possible features for this type of product. Many vendors, for example, most of those participating in this buyers guide, provide one product that covers a wide variety of system monitoring functions; other vendors provide a suite of products each of which cover a range of functions in specific areas. Your needs determine whether you require broad coverage of system events or would be better served by focusing on a specific area.

System monitoring products have become indispensable in environments requiring continuous system availability. If operational problems could disrupt your business operations, you should be running a monitoring product on your AS/400 network to provide advance warning and rapid response.

Three of the most important AS/400 system monitoring product attributes are escalation, notification, and action capabilities. Escalation capabilities let users specify a number of corrective actions to be taken in succession, each more drastic as the problem becomes identified as more severe. For example, an initial "I" (Ignore) response to a message queue error might become a job cancel request or a notification of specific IS department personnel if an error persists or becomes more severe — with the sequence to be followed designated in advance.

Monitoring products’ notification processes support a range of options that include e-mail, digital cellular telephones, and alphanumeric and two-way pagers. In addition, some products let users specify a hierarchy of pagers, telephone numbers, or user IDs to which problem notifications are sent — designations that you can even set to change from shift to shift.

A third important set of functions is action capabilities. Actions can include sending various types of messages to the devices listed above, automatically running CL procedures, responding to system inquiries with preplanned responses, or notifying a list of employees of messages sent and actions taken.

A few term definitions will help you interpret the features table. Interactive job polling for wait status means the product periodically checks to see there are no jobs remaining in wait mode for a specified time period. Administrative central console means the product lets one administrator or operator monitor a network of AS/400s from one workstation. System availability polling means the product checks to see that all AS/400s in a network are communicating with each other, either automatically or on demand. Interrupts operator if problems occur means the product sends a message that is displayed within any operator interactive session in the event of a system anomaly. User-definable status thresholds means users can specify the system performance standards which, if exceeded or failed, cause the product to send a warning message.

Message-filtering capabilities means a product lets users set up barriers that prevent trivial system messages from obscuring important messages, based on criteria as varied as generic job types, specific user or program names, or message severity codes. Filtering by generic job name means you could enter a job name such as "DSP*" and the system would monitor all jobs starting with DSP (e.g., DSP01, DSP02). In contrast, when filtering by specific job name, entering DSP01 would monitor only job DSP01. Generic data comparison means filtering is based on the presence within the message of a generic term such as "urgent," as opposed to specific data comparison, which means the filtering is set up to look for an actual message that’s defined in its entirety.

Under paging support, group paging means messages can be sent to a predesignated group of employees and pager scheduling refers to the attachment of schedules of availability for certain pagers so the product will try to reach them only at certain times. Paging during restricted states means the product can send messages during an IPL or other times when the system is in a restricted state, and pager security means restricting pager messaging authority so only designated users or administrators can send a message to specific pagers.

Dennis Fletcher is a long-time industry specialist in storage and data availability based in Orange County, California. He has been active in COMMON business development sessions, the RAID Advisory Board and the SCSI Trade Assn. He is currently VP, Products at BCC Technologies in Irvine, CA. Dennis can be reached on the Web at http://www.newsrev.com/fletcher .