You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Description
We deploy GPU services using tritonserver 21.12 version. The service adopts a dual docker mode, with one docker for the logical service and another docker for the triton service.
Data communication between GPU and CPU is done through shared memory
Recently, we have found that our service encounters an error like
tritonclient.utils.InferenceServerException: [StatusCode.NOT_FOUND] Unable to find shared memory region: 'data_xx_x'
causing the triton service to be unavailable and all service requests to fail.
After restarting the service, it resumes normal operation.
Triton Information
tritonserver 21.12 version
Are you using the Triton container or did you build it yourself?
using Triton container
The text was updated successfully, but these errors were encountered:
Description
We deploy GPU services using tritonserver 21.12 version. The service adopts a dual docker mode, with one docker for the logical service and another docker for the triton service.
Data communication between GPU and CPU is done through shared memory
Recently, we have found that our service encounters an error like
tritonclient.utils.InferenceServerException: [StatusCode.NOT_FOUND] Unable to find shared memory region: 'data_xx_x'
causing the triton service to be unavailable and all service requests to fail.
After restarting the service, it resumes normal operation.
Triton Information
tritonserver 21.12 version
Are you using the Triton container or did you build it yourself?
using Triton container
The text was updated successfully, but these errors were encountered: