Description of problem: During vagrant box add, after the box was successfully downloaded, I ran out of disk space while the box was unpacking. Vagrant failed as expected with the following error. The box failed to unpackage properly. Please verify that the box [480/895] file you're trying to add is not corrupted and try again. The output from attempting to unpackage (if any): x ./metadata.json x ./Vagrantfile x ./box.img: Write failed bsdtar: Error exit delayed from previous errors. Once I cleaned up my disk space and attempted to add the box again, but it kept failing with: ==> box: Box download is resuming from prior download progress An error occurred while downloading the remote file. The error message, if any, is reproduced below. Please fix this error and try again. The requested URL returned error: 416 Version-Release number of selected component (if applicable): vagrant-1.8.5-3.fc25.noarch How reproducible: deterministic Steps to Reproduce: 1. make enough disk space for box download, but not enough for box unpacking 2. vagrant box add $BOX 3. clean up disk to have enough space for unpacking 4. vagrant box add $BOX Actual results: vagrant fails to add the box after the disk is cleaned up to have enough space Expected results: vagrant successfully adds the box when there's enough disk space Additional info: This upstream issue seems similar and it also mentions a workaround https://github.com/mitchellh/vagrant/issues/4479 Workaround: rm -rf .vagrant.d/tmp
Hi Tomáš, Any chance to reproduce the behavior with either F26+ and Vagrant 1.9.1 or even with the upstream RPM?
Actually, it would be even better if you are able to reproduce the issue with "--debug" flag, since it would give more clue, where the Vagrant failed.
The issue can't be reproduced with vagrant-1.9.1-2.fc26.noarch. If the box add command fails, it removes tmp automatically so the next time the process starts from scratch and works as expected. Feel free to close this for F25, since the workaround is easy.
(In reply to Tomas Krizek from comment #3) > Feel free to close this for F25, since the workaround is easy. Thx for testing. I'm closing it, since this is upstream issue. But please let me know if you by a chance identify specific patch fixing this issue. We might consider backporting it.