Backend.AI is a streamlined, container-based computing cluster platform that hosts popular computing/ML frameworks and diverse programming languages, with pluggable heterogeneous accelerator support including CUDA GPU, ROCm GPU, TPU, IPU and other NPUs.
Is your feature request related to a problem? Please describe.python -m ai.backend.COMPONENT.server --debug makes too many logging messages. (component examples: agent, manager, storage-proxy, web)
Specifically, I experienced log rate spikes with 'address already in use'. Code editors freeze frequently with these types of log floodings.
Describe the solution you'd like
python -m ai.backend.COMPONENT.server --debug would be the default option
There can be workaround options - fatal, error, warn, info, debug, trace.
python -m ai.backend.COMPONENT.server --fatalpython -m ai.backend.COMPONENT.server --warnpython -m ai.backend.COMPONENT.server --error
Combination of the options example.
python -m ai.backend.COMPONENT.server -f -w -e (fatal, warn, error)
(This is just a draft idea. If you have a better opinion, please add a comment!)
Is your feature request related to a problem? Please describe.
python -m ai.backend.COMPONENT.server --debug
makes too many logging messages. (component examples: agent, manager, storage-proxy, web)Specifically, I experienced log rate spikes with 'address already in use'. Code editors freeze frequently with these types of log floodings.
Describe the solution you'd like
python -m ai.backend.COMPONENT.server --debug
would be the default optionThere can be workaround options - fatal, error, warn, info, debug, trace.
python -m ai.backend.COMPONENT.server --fatal
python -m ai.backend.COMPONENT.server --warn
python -m ai.backend.COMPONENT.server --error
Combination of the options example.
python -m ai.backend.COMPONENT.server -f -w -e
(fatal, warn, error)(This is just a draft idea. If you have a better opinion, please add a comment!)
Additional context
Check out this link: https://sematext.com/blog/logging-levels/
The picture shows shells on backend.ai debug mode execution.