i have a long running Jupyter session
i write to an S3 bucket (outside of dotscience) which is mounted as an S3 dataset in my project
i don't want to have to stop and start Jupyter before pulling in the latest changes to the bucket
suggest we have a "synchronize datasets" button in Jupyter, which needs to show status of the long running operation when it's clicked
will need to plumb all the way through from jupyter plugin to communicate with gateway somehow to fetch, then do s3 sync into bucket and then push that dot to the runner while it's mounted (as long as it's mounted readonly, should be fine as zfs recv can write into an active filesystem amazingly)
i have a long running Jupyter session
i write to an S3 bucket (outside of dotscience) which is mounted as an S3 dataset in my project
i don't want to have to stop and start Jupyter before pulling in the latest changes to the bucket
suggest we have a "synchronize datasets" button in Jupyter, which needs to show status of the long running operation when it's clicked
will need to plumb all the way through from jupyter plugin to communicate with gateway somehow to fetch, then do s3 sync into bucket and then push that dot to the runner while it's mounted (as long as it's mounted readonly, should be fine as zfs recv can write into an active filesystem amazingly)