Skip to content

fix: instruction telemetry requires many file descriptors on high core-count systems#673

Open
harp-intel wants to merge 1 commit intomainfrom
fix-processwatch-fd
Open

fix: instruction telemetry requires many file descriptors on high core-count systems#673
harp-intel wants to merge 1 commit intomainfrom
fix-processwatch-fd

Conversation

@harp-intel
Copy link
Contributor

This pull request improves the robustness of the instruction telemetry script by ensuring that the system's open file descriptor limit is high enough to support monitoring on large systems. The main change is the addition of logic to check and, if possible, increase the soft open-file limit before running telemetry. This prevents failures due to "Too many open files" errors when monitoring many CPUs.

Resource limit handling:

  • Added logic to calculate the required open file descriptor limit based on the number of CPUs, ensure the current limit is sufficient, and attempt to raise it if needed. If the limit cannot be raised, the script now exits with an informative error message. (internal/script/scripts.go) [1] [2]

…stems

Signed-off-by: Harper, Jason M <jason.m.harper@intel.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant