vLLM versions prior to 0.14.1 are deprecated due to security issue CVE-2026-22778 which can allow remo

cuda-gdb segmentation fault on startup

Resolution: 
Resolved

The CUDA debugger, cuda-gdb, can raise a segmentation fault immediately upon execution.  A workaround before executing cuda-gdb is to unload the xalt module, e.g.: 

module unload xalt

This issue affects most cuda modules on Pitzer and Owens.