Machine learning container keeps crashing after upgrade to 1.62.1
After upgrading to 1.62.1 the machine learning container keeps crashing. The logs are attached
How do I fix this?
5 Replies
Try deleting the model cache and restarting the container
looks like it worked. does the cache need to be cleaned before an upgrade?
Not necessarily, though it probably can't hurt. We only see this error occasionally
ok
that's an odd error. I checked the poetry.lock and the protobuf version hasn't changed.
it could be caused by the model download being interrupted, like if the container restarted or something.
if it happens again it'd be great if you could upload the model so i can check if it's corrupt or if it's something else