java: Paranoid profiling stop on suspicious events in kernel messages #235
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
We want the ability to be very strict on error detection. This is possibly too strict (segfaults in non-profiled processes are most likely not related to us...) but we'll start very strict and lower the bar as we go.
The "PID in message" check will help us identify if our regex on OOM/signals miss anything (or if there's any other type of bad message we should be looking at). I hope it won't catch too many false positives.
How Has This Been Tested?
docker run -it -m 6m pythonand made sure Java profiling stopsmain(){*(int*)0=0;}and made sure Java profiling stopsecho 374075 | sudo tee /dev/kmsg) and made sure Java profiling stops. It's funny that it works - the kernel does flag messages coming from usermode, we should check this flag and ignore such messages as we really care only about kernel messages.