WangGuangxin opened a new pull request, #11676: URL: https://github.com/apache/incubator-gluten/pull/11676
## What changes are proposed in this pull request? In Vanilla Spark WebUI, it is easy to get the current java jstack, which is very helpful to diagnostic the hang/slow problems. But when it comes to Spark on Gluten, the jstack is meaningless since all heavy jobs are under JNI call. This PR adds the same ability to capture native C++ stack traces from executors directly from the Gluten SQL tab in the Spark UI. It introduces an executors overview with a per‑executor "C++ Stack" action and a dedicated "Gluten C++ Stack" status page that streams the collected stack output. <img width="3072" height="1208" alt="image" src="https://github.com/user-attachments/assets/3cea55b8-13b4-4971-b427-d0858cd75d73"; /> <img width="2490" height="1118" alt="image" src="https://github.com/user-attachments/assets/1df4f0af-2d91-4ba2-9afd-8438af50bfca"; /> Note that current implementation is asynchronous, mainly because gdb data collection typically takes a long time, exceeding the yarn proxy timeout in our environment. ## How was this patch tested? Manually ## Was this patch authored or co-authored using generative AI tooling? Generated-by: TRAE with Doubao-Seed-Code -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
