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
The other day I was trying to install OPEA ChatQnA example on a machine with Intel Xeon CPU. After following all the steps detailed in the documentation, the UI interface worked on my browser. However the chat requests failed with messaged "failed to load chat content". On inspecting the docker containers that are running, I noticed three of the microservices viz:- vllm, redis and tei had exited due to some error. further debugging why those containers exited i found there was a error during the service launch "api_server.py: error: argument --model: expected one argument".
I had done ChatQnA installation several times in the past on different Xeon machines and never faced this issue. On extensive debugging we were finally able to figure out that when export the environment variables with "source ./set_env.sh" and other export functions, I didn't use sudo (it was not required). However on this particular machine, running docker compose command needed sudo privilege. I had forgotten that exporting environment variables as normal user while executing docker compose with root made the environment variables to be blank or unset for docker compose.
So if your system requires that docker commands run on sudo, then please ensure environment variables are set with root as well. I switched to root with "sudo su -p" and then exported environment variables and executed further steps. This resolved my issue. if anyone else gets this issue, hope this helps.
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
The other day I was trying to install OPEA ChatQnA example on a machine with Intel Xeon CPU. After following all the steps detailed in the documentation, the UI interface worked on my browser. However the chat requests failed with messaged "failed to load chat content". On inspecting the docker containers that are running, I noticed three of the microservices viz:- vllm, redis and tei had exited due to some error. further debugging why those containers exited i found there was a error during the service launch "api_server.py: error: argument --model: expected one argument".
I had done ChatQnA installation several times in the past on different Xeon machines and never faced this issue. On extensive debugging we were finally able to figure out that when export the environment variables with "source ./set_env.sh" and other export functions, I didn't use sudo (it was not required). However on this particular machine, running docker compose command needed sudo privilege. I had forgotten that exporting environment variables as normal user while executing docker compose with root made the environment variables to be blank or unset for docker compose.
So if your system requires that docker commands run on sudo, then please ensure environment variables are set with root as well. I switched to root with "sudo su -p" and then exported environment variables and executed further steps. This resolved my issue. if anyone else gets this issue, hope this helps.
Beta Was this translation helpful? Give feedback.
All reactions