The Lounge is rated Safe For Work. If you're about to post something inappropriate for a shared office environment, then don't post it. No ads, no abuse, and no programming questions. Trolling, (political, climate, religious or whatever) will result in your account being removed.
It isn't that, it is the build trying to do too much at once, spawning processes everywhere. Apparently, and rather ridiculously, it seems to be incapable of adapting dynamically to the resource availability of the host machine!
That's a great idea! If you are caught by an access violation, preventing you from destroying data that isn't yours, obtain the rights to do the write so that the other data IS overwritten. If you are lucky, that might bring down your entire system, not just the worker process!
Are you sure? Might it be that this is just another case that you disagree with what is written in the manual and so choose to visit Code Project to have a rant about it instead.
It could be that error messages are made up there are no error messages, all these people have been wrong about error messages for years, there is evidence to support that error messages are not real and if we all just ignore them everything will be fine in the end.
My programs (/processes) always have an outermost exception handler catching all otherwise uncaught exceptions. My support library provides a ComeFrom function used by exception handlers to traverse the stack to pick up whatever information is available.
True enough: There is a standard "Print stacktrace" option. But first: Handling the exception programmatically is far easier when the information is provided as a data structure, rather than in console output format. Second, there is often more information available in the binary stack traversal than what is printed on the console.
If the code is not yours, so you cannot insert an exception handler, this option may be unavailable. But in those cases, you often cannot do anything about it anyway. (Our Bamboo build system generates at least a handful of stack traces every day, often 250-300 stack frames deep. We can read the function names, so what? We can do nothing but ignore it. Mailing the stack trace to Atlassian is futile.)