SZEDER Gábor [off-list ref] 于2021年1月20日周三 下午8:29写道：

On Mon, Jan 18, 2021 at 04:24:11PM +0800, Jiang Xin wrote:

quoted

SZEDER Gábor [off-list ref] 于2021年1月18日周一 上午6:21写道：

quoted

This patch added a whole lot of test cases, and two of them '86 -
proc-receive: not support push options (builtin protocol)' and '95 -
proc-receive: not support push options (builtin protocol/porcelain)'
are prone to rare failures.

On Thu, Aug 27, 2020 at 11:45:44AM -0400, Jiang Xin wrote:

quoted

diff --git a/t/t5411/test-0026-push-options.sh b/t/t5411/test-0026-push-options.sh
new file mode 100644
index 0000000000..d0c4da8b23
--- /dev/null
+++ b/t/t5411/test-0026-push-options.sh

quoted

+# Refs of upstream : master(A)
+# Refs of workbench: master(A)  tags/v123
+# git push -o ...  :                       refs/for/master/topic
+test_expect_success "proc-receive: not support push options ($PROTOCOL)" '
+     test_must_fail git -C workbench push \
+             -o issue=123 \
+             -o reviewer=user1 \
+             origin \
+             HEAD:refs/for/master/topic \
+             >out 2>&1 &&

Three relevant things are happening here:

  - 'git push' is executed with its standard output and error
    redirected to the file 'out'.

  - 'git push' executes 'git receive-pack' internally, which inherits
    the open file descriptors, so its output and error goes into that
    same 'out' file.

  - 'git push' is expected to fail when it finds out that the other
    side doesn't support push options, but it does so with a simple
    die() right away, without waiting for its child 'git receive-pack'
    process to finish.

quoted

+     make_user_friendly_and_stable_output <out >actual &&
+     test_i18ngrep "fatal: the receiving end does not support push options" \
+             actual &&
+     git -C "$upstream" show-ref >out &&

Here the shell opens and truncates the file 'out' to write 'git
show-ref's output, i.e. it is still the same 'out' file that was used
earlier.

Consequently, it is possible that 'git receive-pack' is still running,
its open file descriptors to 'out' are still valid, and its "fatal:
the remote end hung up unexpectedly" error message about the suddenly
disappeared 'git push' can partially overwrite the output from 'git
show-ref'.

quoted

quoted

I think these are the only two tests that can cause this racy
behavior: by instrumenting finish_command() I found that in all other
tests where 'git push' is expected to fail it errors out gracefully
and waits for its 'git receive-pack' child process.

Atomic push may have the same problem.

I don't think so, because send_pack() doesn't die() when a ref is
rejected in an atomic push, but returns, and lets its caller terminate
in an usual way, including waiting for 'git receive-pack'.

I find many places where the client side will die() before closing the
service side gracefully:

 + In `transport_push()`, if fail to push a submodule, will die().
 + In `git_transport_push()`, will die() for an unimplemented v2 protocol.
 + In `send_pack()`, will die() if
    the server side has an incompatible hash algorithm, or
    the receiving end does not support --signed push, or
    the receiving end does not support --atomic push, or
    the receiving end does not support push options