Error 134 happening consistently on one worker only
I have 9 serverless workers for my endpoint but consistently get the message "[error]worker exited with exit code 134" on only one of them. This seems to happen after loading the model, which I presume means out-of-memory or something, which surprises me since it is on 4090 and the model isn't that big.
The endpoint id is 9afi4omg7sdwt6 and the worker that is failing is mqz3hcdwun2i1i.
I have terminated the worker because I can't keep having this problem.
The endpoint id is 9afi4omg7sdwt6 and the worker that is failing is mqz3hcdwun2i1i.
I have terminated the worker because I can't keep having this problem.
