From a02ea577174ab8ed18f847cf1693f213e0b9c473 Mon Sep 17 00:00:00 2001 From: Jeff King Date: Thu, 7 Jan 2021 04:43:58 -0500 Subject: [PATCH 1/2] git_connect_git(): forbid newlines in host and path When we connect to a git:// server, we send an initial request that looks something like: 002dgit-upload-pack repo.git\0host=example.com If the repo path contains a newline, then it's included literally, and we get: 002egit-upload-pack repo .git\0host=example.com This works fine if you really do have a newline in your repository name; the server side uses the pktline framing to parse the string, not newlines. However, there are many _other_ protocols in the wild that do parse on newlines, such as HTTP. So a carefully constructed git:// URL can actually turn into a valid HTTP request. For example: git://localhost:1234/%0d%0a%0d%0aGET%20/%20HTTP/1.1 %0d%0aHost:localhost%0d%0a%0d%0a becomes: 0050git-upload-pack / GET / HTTP/1.1 Host:localhost host=localhost:1234 on the wire. Again, this isn't a problem for a real Git server, but it does mean that feeding a malicious URL to Git (e.g., through a submodule) can cause it to make unexpected cross-protocol requests. Since repository names with newlines are presumably quite rare (and indeed, we already disallow them in git-over-http), let's just disallow them over this protocol. Hostnames could likewise inject a newline, but this is unlikely a problem in practice; we'd try resolving the hostname with a newline in it, which wouldn't work. Still, it doesn't hurt to err on the side of caution there, since we would not expect them to work in the first place. The ssh and local code paths are unaffected by this patch. In both cases we're trying to run upload-pack via a shell, and will quote the newline so that it makes it intact. An attacker can point an ssh url at an arbitrary port, of course, but unless there's an actual ssh server there, we'd never get as far as sending our shell command anyway. We _could_ similarly restrict newlines in those protocols out of caution, but there seems little benefit to doing so. The new test here is run alongside the git-daemon tests, which cover the same protocol, but it shouldn't actually contact the daemon at all. In theory we could make the test more robust by setting up an actual repository with a newline in it (so that our clone would succeed if our new check didn't kick in). But a repo directory with newline in it is likely not portable across all filesystems. Likewise, we could check git-daemon's log that it was not contacted at all, but we do not currently record the log (and anyway, it would make the test racy with the daemon's log write). We'll just check the client-side stderr to make sure we hit the expected code path. Reported-by: Harold Kim Signed-off-by: Jeff King Signed-off-by: Junio C Hamano --- connect.c | 2 ++ t/t5570-git-daemon.sh | 5 +++++ 2 files changed, 7 insertions(+) diff --git a/connect.c b/connect.c index 79f1b3b242..7b4b65751d 100644 --- a/connect.c +++ b/connect.c @@ -1063,6 +1063,8 @@ static struct child_process *git_connect_git(int fd[2], char *hostandport, target_host = xstrdup(hostandport); transport_check_allowed("git"); + if (strchr(target_host, '\n') || strchr(path, '\n')) + die(_("newline is forbidden in git:// hosts and repo paths")); /* * These underlying connection commands die() if they diff --git a/t/t5570-git-daemon.sh b/t/t5570-git-daemon.sh index 7466aad111..336d417a90 100755 --- a/t/t5570-git-daemon.sh +++ b/t/t5570-git-daemon.sh @@ -102,6 +102,11 @@ test_expect_success 'fetch notices corrupt idx' ' ) ' +test_expect_success 'client refuses to ask for repo with newline' ' + test_must_fail git clone "$GIT_DAEMON_URL/repo$LF.git" dst 2>stderr && + test_i18ngrep newline.is.forbidden stderr +' + test_remote_error() { do_export=YesPlease From 6aed56736b882f94c81293d1646d27d10241349c Mon Sep 17 00:00:00 2001 From: Jeff King Date: Thu, 7 Jan 2021 04:44:17 -0500 Subject: [PATCH 2/2] fsck: reject .gitmodules git:// urls with newlines The previous commit taught the clone/fetch client side to reject a git:// URL with a newline in it. Let's also catch these when fscking a .gitmodules file, which will give an earlier warning. Note that it would be simpler to just complain about newline in _any_ URL, but an earlier tightening for http/ftp made sure we kept allowing newlines for unknown protocols (and this is covered in the tests). So we'll stick to that precedent. Signed-off-by: Jeff King Signed-off-by: Junio C Hamano --- fsck.c | 2 +- t/t7416-submodule-dash-url.sh | 15 +++++++++++++++ 2 files changed, 16 insertions(+), 1 deletion(-) diff --git a/fsck.c b/fsck.c index c4a9fe624a..a8870a54b6 100644 --- a/fsck.c +++ b/fsck.c @@ -1086,7 +1086,7 @@ static int check_submodule_url(const char *url) if (looks_like_command_line_option(url)) return -1; - if (submodule_url_is_relative(url)) { + if (submodule_url_is_relative(url) || starts_with(url, "git://")) { char *decoded; const char *next; int has_nl; diff --git a/t/t7416-submodule-dash-url.sh b/t/t7416-submodule-dash-url.sh index eec96e0ba9..d21dc8b009 100755 --- a/t/t7416-submodule-dash-url.sh +++ b/t/t7416-submodule-dash-url.sh @@ -201,4 +201,19 @@ test_expect_success 'fsck rejects embedded newline in relative url' ' grep gitmodulesUrl err ' +test_expect_success 'fsck rejects embedded newline in git url' ' + git checkout --orphan git-newline && + cat >.gitmodules <<-\EOF && + [submodule "foo"] + url = "git://example.com:1234/repo%0a.git" + EOF + git add .gitmodules && + git commit -m "git url with newline" && + test_when_finished "rm -rf dst" && + git init --bare dst && + git -C dst config transfer.fsckObjects true && + test_must_fail git push dst HEAD 2>err && + grep gitmodulesUrl err +' + test_done