Cloning LFS sometimes fails with read tcp i/o timeout

[Miro Knejp]

For about a week or so we have sporadic failures on our SppVeyor server when cloning our repository. When it happens the message is always the same

batch response: Post https://bitbucket.org/user/repo.git/info/lfs/objects/batch: read tcp 1.2.3.4:1051->5.6.7.8:443: i/o timeout

We're cloning the repository with

git lfs clone --branch=foo https://bitbucket.org/user/repo

The version of git on AppVeyor is currently 2.13.0(2) and git lfs is 2.1.0

Is there anything that can be done to check whether this is a problem on Bitbucket or AppVeyor? It only started occurring about 5 days ago and since then it breaks our builds multiple times a day because checked out binary files end up corrupted. I couldn't find a way to run something like an LFS integrity check on the bitbucket interface.

[Edwin Kyalangalilwa]

Try a regular

git clone --branch=foo https://bitbucket.org/user/repo

It will download the repo first then download the binary files