Metrics API doesn't report all three default metrics in case of GRPC

[thies1006]

Hello! I noticed that when sending requests to my server via GRPC the two metrics ts_inference_latency_microseconds counter, ts_queue_latency_microseconds counter are not reported via metrics API (curl http://IP:PORT/metrics). However I do see ts_inference_requests_total counter.

Now when sending the requests via REST I get all three reported.

• torchserve version: 0.5.2
• torch-model-archiver version: 0.5.2
• torch version: 1.10.0
• torchvision version [if any]:
• torchtext version [if any]:
• torchaudio version [if any]:
• java version: 11.0.6+10
• Operating System and version: ubuntu 18.04 LTS

Your Environment

• Installed using source? no:
• Are you planning to deploy it using docker container? [yes/no]: no
• Is it a CPU or GPU environment?: GPU
• Using a default/custom handler? default nmt handler
• What kind of model is it e.g. vision, text, audio?: text
• Are you planning to use local models from model-store or public url being used e.g. from S3 bucket etc.? [If public url then provide link.]:
• Provide config.properties, logs [ts.log] and parameters used for model registration/update APIs:
• Link to your project [if any]:

Expected Behavior

get all three default metrics via GRPC

[IamMohitM]

I'm facing the same issue. No ts_inference_latency_microseconds and ts_queue_latency_microseconds metrics.

Using torchserve version 0.6.0

[harshita-meena]

I am facing the exact same issue with gRPC, only the count of requests are exported to the metrics API but not the latency. Wondering what is the status of the bug fix.