openvino repo GPG key expired

[jdesfossez]

Using the mlperf-inference-v2.0-cpu or mlperf-inference-v2.0 of the triton-inference-server repository causes it to try to install an openvino repository with an expired GPG key:
https://github.com/triton-inference-server/server/blob/mlperf-inference-v2.0-cpu/qa/common/gen_qa_model_repository#L193

root@03e4af619db3:/# wget https://apt.repos.intel.com/openvino/2021/GPG-PUB-KEY-INTEL-OPENVINO-2021
[...]
2023-10-03 17:38:56 (310 MB/s) - 'GPG-PUB-KEY-INTEL-OPENVINO-2021' saved [939/939]
root@03e4af619db3:/# apt-key add GPG-PUB-KEY-INTEL-OPENVINO-2021 
OK
root@03e4af619db3:/# apt-key list
/etc/apt/trusted.gpg
--------------------
pub   rsa2048 2019-09-30 [SC] [expired: 2023-09-30]
      52AB D6E8 7E42 1793 9718  73FF ACFA 9FC5 7E6C 5DBE
uid           [ expired] Intel(R) Software Development Products

This is the code triggering this call:

inference_results_v3.1/closed/NVIDIA/scripts/build_triton_cpu_libs.sh

Line 21 in eaf622a

build_branch="mlperf-inference-v2.0-cpu"

The newer releases of the triton-inference-server use a different method to install openvino.
Should that branch be updated or should that issue be raised to openvino ?

[mrasquinha-g]

@nvyihengz Can someone from Nvidia take a look at this issue.

[nvzhihanj]

Hi, Triton-inference-server and openvino are not used for submission by NVIDIA in 3.1. This is a stale script that's not removed - we will update the repo in v4.0.