-
Notifications
You must be signed in to change notification settings - Fork 573
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鈥檒l occasionally send you account related emails.
Already on GitHub? Sign in to your account
bug: 503 for /readyz with model-id facebook/opt-125m #79
Comments
We updated our packages with |
I have addressed this on the main. It leads down to an internal bug in how I first implemented the load model in the bento. I will release a new patch |
Can you try with 0.1.17? |
We upgraded to 0.1.17, but the behavior did not change. |
Thank you for your effort btw 馃挴 |
After enabling Debug mode, we found out that the browser seems to replace the colons in the URL leading to a 404 status code: Is there anything else I can do to help? 馃樃 |
Hi @aarnphm, we upraded to version openllm-0.1.20. Currently the bug seems to be fixed, we get an status code 200 from the /readyz endpoint. We will further investigate the other endpoints as well. 馃惐 |
I don't know why, but it does not work again. Meanwhile, we tried other models but we did not change any config at all. /readyz says again
|
can you walk me through how you run this again? are you just doing |
What is the resource you are running on? |
Yeah, its a virtual environment and we Run |
Its a virtual machine running on VMware esxi, without GPU, it is CPU only |
i can successfully run opt without any hiccups on mac. OPT shouldn't require GPU to run at all |
Hi, we found out it was the proxy who caused the issue. We needed to add |
Describe the bug
First at all: Thank you very much, openllm looks awesome so far 馃挴
This issue is regarding to #47. We tried to start an openllm server with the command:
openllm start opt --model-id facebook/opt-125m
The server started successfully and the webinterface is reachable. But we cannot generate anything, the on the webinterface given examples do not work.
openllm query
did not work either.To reproduce
openllm start opt --model-id facebook/opt-125m
Logs
System information
bentoml
: 1.0.22python
: 3.9.2platform
: Linux-4.18.0-305.76.1.el8_4.x86_64-x86_64-with-glibc2.28uid_gid
: 8007:8008pip_packages
transformers
version: 4.30.2System information (Optional)
memory: 240 GB
CPU: 16 vCPU
Platform: VMWare ESXi 7.0 U 3
OS: RHEL 8.4
The text was updated successfully, but these errors were encountered: