As a user of AIMMS applications in a large scale federated energy modeling framework, using AIMMS as endpoint, hundreds of times a night,
I would like to manage the logging of AIMMS errors or messages on a per-request basis by clearing errors and messages in the beginning of each request so that a relevent copy of .err and log files produced during a API task can be saved with a particular request.
The value of this user story is easier management of persisted logging information,
including what logging can be safely cleared, and only relevant information gathering on particular API tasks warrant a root cause analysis can be collected.
Page 1 / 1
New→Discussion ongoing
I'm not particularly sure that the .err and .log files offer the logging quality needed to do a thorough root cause analysis of any problems in your application, and ultimately whether we should make an effort to split the log and error files. The .err will register every AIMMS error or warning regardless of whether it's handled or not, the .log file will only spit out all AIMMS errors and warning not handled.
Doing dedicated logging at the application level where next to AIMMS errors and warnings you also log data and optimization problems in your model and add tracing to follow the task's progress, seems like a more useful approach to enable root cause analysis. You can then store that dedicated application log with the task results, which will give you a much better overview what happened during the execution of the task.
Sign up
Already have an account? Login
Please use your business or academic e-mail address to register
Login to the community
No account yet? Create an account
Enter your E-mail address. We'll send you an e-mail with instructions to reset your password.