Replies: 1 comment
-
Need to investigate how crash dumps generated in eBPF binaries in production can be consumed for investigation. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Describe the bug
We currently have 'asserts' for un-recoverable and/or fatal errors. These asserts are triggered in debug builds but are compiled out in 'release' builds.
This can often lead to silent failures and/or buggy system behavior in release builds. Also, considering that most ebpf consumers would prefer running the retail builds (ebpf program verification times are un-acceptably slow in debug builds), the subsequent inconsistent behavior would only end up confusing the ebpf end customers.
We should bugcheck the system and generate a crash dump to debug and fix issues such as:
OS information
No response
Steps taken to reproduce bug
N/A
Expected behavior
The system should bugcheck and generate a crash dump.
Actual outcome
The failures are ignored and system behavior becomes indeterminate.
Additional details
No response
Beta Was this translation helpful? Give feedback.
All reactions