-
Notifications
You must be signed in to change notification settings - Fork 228
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[BUG] Failed to split an empty string with error "ai.rapids.cudf.CudfException: parallel_for failed: cudaErrorInvalidDevice: invalid device ordinal" #11183
Comments
This is due to a bug in cudf: rapidsai/cudf#16284. |
@LIN-Yu-Ting is the query using a string split operation as described in this issue? If not, it would be best to file this as a new issue, as the root cause is likely to be very different. Also, if you have a way for us to reproduce the issue you're seeing, that would be a big help. |
I see that the exception is thrown from |
Yes, that's definitely different than the string split originally reported here. However I'm wondering if the chunked reader error was the result a "sticky" CUDA error that actually occurred elsewhere first. If this is the only error being reported in the executor logs then that would be a different issue than the one reported here. |
Closing this issue since it is already fixed in 24.08 branch |
Describe the bug
Query failed to split an empty string with error :
Steps/Code to reproduce bug
Expected behavior
CPU mode works fine and we should follow the same behavior:
Environment details (please complete the following information)
Dataproc 2.1
Spark RAPIDS 24.08 snapshot jar
The text was updated successfully, but these errors were encountered: