In the tag-and-push.sh script, if the tagging fails, (because :latest wasn't build properly), but 1.0 still existed, that old version would get pushed to the Hub. One of my students ran into this, and it took us a while to figure out why what was happening.
It's best practice to "set -e" so we immediately throw an error if the tagging fails, so we don't upload any old images.
In the tag-and-push.sh script, if the tagging fails, (because
:latestwasn't build properly), but 1.0 still existed, that old version would get pushed to the Hub. One of my students ran into this, and it took us a while to figure out why what was happening.It's best practice to "set -e" so we immediately throw an error if the tagging fails, so we don't upload any old images.