Before the Web was an everyday resource for legacy managers, list servers did the serving-up of tech intelligence. You’d sign up for a regular mailing of traffic between experts and managers. Every day, for the busiest lists, your email box would fill with messages about how to resolve system issues like the one below, for MPE/iX servers and their ImageSQL databases
We have a bunch of old Image log files having the form HLHAM999 where 999 appears to be a serial number. Unfortunately we seem to have used up all the numbers (001-999) and the logging process seems stuck on using a single file (HLHAM899). So my question is, does the logging process automatically roll over and reuse old log files, or do I have to purge these things myself?
Since 1994, those questions and tens of thousands of answers have lived on a server in Tennessee. The University of Tennessee at Chattanooga (UTC) had some of the earliest-installed HP 3000 servers among the academic community. Jeff Kell, the 3000 administrator with powerful networking skills, put up a Listserv, as it was called, to serve MPE/iX needs.
The good news is that the database is searchable, even today. The UTC server will locate messages using search strings that include subject line. The searches can even locate answers within date ranges. That’s especially important to a legacy community like the one for the HP 3000, where even answers from 20 years ago have relevance to the systems of today.
Even in 2016, an HP 3000 manager who was dusting off an archival server had questions about a database and its logging needs. “If I must perform a manual purge, then how to I handle the logging process? Do I stop it and purge everything? How to I set the serial number for the logger when I restart it? In fact, how do I turn the logger off and on if that proves necessary?
Other managers chip in to answer questions like there, or the independent support community provides solutions. It’s still running today, just like the legacy servers in the HP, VMS, Tru64 and Solaris communities.