tar: This does not look like a tar archive

tar: This does not look like a tar archive means the file read by tar -x has no valid tar header. The file is not what its name says: an HTML error page, a truncated download or a different archive format. Inspect it with file before changing tar options.

Reproduce it

echo '<html>404 Not Found</html>' > download.tar
tar -xf download.tar

tar: This does not look like a tar archive
tar: Exiting with failure status due to previous errors

With GNU tar the first line is followed by tar: Exiting with failure status due to previous errors, and the exit status is 2.

Check what the file is

echo '<html>404 Not Found</html>' > download.tar
file download.tar
head -c 40 download.tar

download.tar: HTML document, ASCII text
<html>404 Not Found</html>

A real archive gives a different answer:

mkdir src && echo hello > src/a.txt
tar -czf good.tar.gz src
file good.tar.gz | cut -d, -f1
tar -tf good.tar.gz

good.tar.gz: gzip compressed data
src/
src/a.txt

curl -f makes HTTP errors fail instead of saving the error page, and -L follows redirects. For .tar.gz problems with gzip see gzip: stdin: not in gzip format.