Using the docker container to run eland_import_hub_model and install a large model imposes large memory requirements. For example using the command below to install xlm-roberta-base requires the container to have more than 8GB of memory.
If the container does not have enough memory the process exits shortly after the download has completed prior to uploading the model to Elasticsearch. Investigate the what is causing the high memory usage.
What is causing the high memory usage?
Tracing the mode?
How can a failure due to insufficient memory be detected and reported?
Using the docker container to run
eland_import_hub_model
and install a large model imposes large memory requirements. For example using the command below to install xlm-roberta-base requires the container to have more than 8GB of memory.If the container does not have enough memory the process exits shortly after the download has completed prior to uploading the model to Elasticsearch. Investigate the what is causing the high memory usage.