You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
WORKDIR /tmp
RUN wget -q http://www.us.apache.org/dist/kafka/${KAFKA_VERSION}/${KAFKA_RELEASE_ARCHIVE}
RUN wget -q https://dist.apache.org/repos/dist/release/kafka/${KAFKA_VERSION}/${KAFKA_RELEASE_ARCHIVE}.md5
The text was updated successfully, but these errors were encountered:
Hi, thanks for reporting, but I have not observed this issue—take for example the latest automated build on Docker Hub where you'll see in the build log output that the checksum is indeed being computed/printed from the tgz.
As the article you linked says,
Interestingly, the URL download and archive unpacking features cannot be used together. Any archives copied via URL will NOT be automatically unpacked.
So unless something has changed quite recently in Docker, there should not be premature unpacking occurring. Perhaps you can tell me if there is any more to reproducing the problem? I see an additional step for modifying java.security in your output so perhaps you have some modifications.
I'm not so bothered by the concerns of "magic" that article takes an opinionated stance on, personally. There's a reasonable argument there for curl/wget to pipeline commands and reduce layers, but that doesn't apply if I want to keep the step of saving the archive on disk to verify its digest.
Because of the "magic" ADD, your tgz is being unpacked prematurely. see https://www.ctl.io/developers/blog/post/dockerfile-add-vs-copy/. This means that there is no file for the checksum to verify. Replace ADD with WGET...
I'm suggesting you replace those with...
The text was updated successfully, but these errors were encountered: