On Thu, Feb 21, 2019 at 12:24:41PM -0800, Jonathan Tan wrote: > When transmitting and receiving POSTs for protocol v0 and v1, > remote-curl uses post_rpc() (and associated functions), but when doing > the same for protocol v2, it uses a separate set of functions > (proxy_rpc() and others). Besides duplication of code, this has caused > at least one bug: the auth retry mechanism that was implemented in v0/v1 > was not implemented in v2. > > To fix this issue and avoid it in the future, make remote-curl also use > post_rpc() when handling protocol v2. Because line lengths are written > to the HTTP request in protocol v2 (unlike in protocol v0/v1), this > necessitates changes in post_rpc() and some of the functions it uses; > perform these changes too. This patch seems to cause noop fetches from GitHub to report "the remote end hung up unexpectedly" at the end of the fetch (I have protocol.version=2 set in the examples below): [parent is good] $ git checkout 0cdb2a12ad0300b5d0cb5bb6e8999034ae4b9bef^ $ make $ bin-wrappers/git fetch https://github.com/git/git From https://github.com/git/git * branch HEAD -> FETCH_HEAD [this patch is bad] $ git checkout 0cdb2a12ad0300b5d0cb5bb6e8999034ae4b9bef $ make $ bin-wrappers/git fetch https://github.com/git/git From https://github.com/git/git * branch HEAD -> FETCH_HEAD fatal: the remote end hung up unexpectedly I haven't dug, so it's possible that GitHub's server-side is doing something slightly odd or wrong. But given the nature of the patch, I'd suspect we just aren't noticing EOF correctly. -Peff
