Originally posted by: repanzer1
I can not recall any time where one map in system was not running and the other maps in that system were active. Usually when there is a problem in one map, like a fatal error that stops it from running, it brings down the entire system. Other systems can be up and running while one system is down, but I do not believe you'll ever have a map have a serious problem and launcher stops executing it, while it still allows other maps in the system to be active. When you bring up a system, the maps in that system are read into memory for one process with one process ID under Linux, so if something causes a fatal error in the process, it should effect the entire process, not allow some other items in there to still be active. Of course, this is an educated assumption. I can only go by what I know, as I'm not a Linux expert, and can only offer my experience with WTX, and in my experience, I have never seen that situation. I will say that if it were my shop and I had to make this determination, I code under the assumption that if one map is up, they all are up.
Actually, now that I think about it, in my last 2 jobs, the way were knew a system was not up, is each system had a map that had one output card which wrote a file every 2 minutes or so, and that file just said “System 1 is up”. A batch file, whether it was under Linux, Unix or Windows, would read that file and if it was not updated in the last two minutes, a beep would go out to the support beeper, so we relied on the fact that if one map had a problem, the entire system went down, else this above process would not work. It never failed us. I think you’re good.
There are 10 types of people in this world. Those who understand binary and those who don't.
#DataExchange#IBMSterlingTransformationExtender#IBM-Websphere-Transformation-Extender