We have products that monitor system log messages. I'd like to know how to do that. In fact, a way to monitor any started task's JESMSGLG would be a real boon. Anybody know how to do that? Thanks.
Thanks for your response. No, the size of the log is not what I want to monitor. What I'd like to monitor are the messages, themselves, so that I can respond to conditions as they occur.
I think it would be helpful to everyone if you gave a bit more information about your system environment.
If you are using a mainframe of somesort, then the system documentation will most likely describe how the system broadcasts its messages - giving you clues on where to look.
An example. I am using a Tandem system (Guardian OS). The operating system automatically sends its messages to a special process. This process has a reserved name, which when it exists (the implementer decides to name their user written running program after this special process name), the OS sends it its messages. Programmers name their running code after this process in order to do exactly what you want to do.
For a different solution, I think TonHu is quite right in that, if you are reading (monitoring) the log file, a change in its size will tell you that there is new data to process - advance your record pointer, et voila! Good idea, indeed.
I have, in the past, used TonHu's suggestion on my system. The only difference was that my OS comes with a set of APIs that you can use when you need to be alerted to the presence of new data in a monitored file.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.