From: Randall S. Becker <rsbecker@nexbridge.com>
Date: Fri, 08 Feb 2019 20:38:05 GMT
Subject: RE: [Breakage] Git v2.21.0-rc0 - t5318 (NonStop)
Message-ID: <003a01d4bfee$349245b0$9db6d110$@nexbridge.com>
In-Reply-To: <20190208193157.GA30952@sigill.intra.peff.net>



> -----Original Message-----
> From: Jeff King <peff@peff.net>
> Sent: February 8, 2019 14:32
> To: Randall S. Becker <rsbecker@nexbridge.com>
> Cc: 'Junio C Hamano' <gitster@pobox.com>; git@vger.kernel.org; 'Linux
> Kernel' <linux-kernel@vger.kernel.org>; git-packagers@googlegroups.com
> Subject: Re: [Breakage] Git v2.21.0-rc0 - t5318 (NonStop)
> 
> On Fri, Feb 08, 2019 at 02:26:17PM -0500, Randall S. Becker wrote:
> 
> > > > For this, we could use truncate -s count file instead of dd to get
> > > > a fixed size file of nulls. This would remove the need for
> > > > /dev/zero in
> > > > t5318 (the patch below probably will wrap badly in my mailer so I
> > > > can submit a real patch separately.
> > >
> > > I don't think "truncate" is portable, though.
> >
> > It is available AFAIK on Linux, POSIX, and Windows under Cygwin.
> > That's more than /dev/zero has anyway. I have the patch ready if you
> > want it.
> 
> Is it POSIX? Certainly truncate() is, but I didn't think the command-line tool
> was. If it really is available everywhere, then yeah, I'd be fine with it.
> 
> > > > > Other cases don't seem to actually care that they're getting
> > > > > NULs, and are just redirecting stdin from /dev/zero to get an
> > > > > infinite amount of input. They could probably use "yes" for that.
> > > >
> > > > What about reading from /dev/null?
> > >
> > > That would yield zero bytes, not an infinite number of them.
> >
> > So something like: yes | tr 'y' '\0' | stuff?
> 
> Exactly (if we even care about them being NULs; otherwise, we can omit the
> "tr" invocation).

I'm a bit perplexed about this... Obviously added some debugging info, but why we're getting No REQUEST_METHOD is perplexing. Is this a lack of an apache2 instance?

expecting success:
        NOT_FIT_IN_SSIZE=$(ssize_b100dots) &&
        env \
                CONTENT_TYPE=application/x-git-upload-pack-request \
                QUERY_STRING=/repo.git/git-upload-pack \
                PATH_TRANSLATED="$PWD"/.git/git-upload-pack \
                GIT_HTTP_EXPORT_ALL=TRUE \
                REQUEST_METHOD=POST \
                CONTENT_LENGTH="$NOT_FIT_IN_SSIZE" \
                yes | tr "y" "\\0" | git http-backend 2>err &&
        echo "Err is" &&
        cat err &&
        grep "fatal:.*CONTENT_LENGTH" err

Status: 500 Internal Server Error
Expires: Fri, 01 Jan 1980 00:00:00 GMT
Pragma: no-cache
Cache-Control: no-cache, max-age=0, must-revalidate

Err is
fatal: No REQUEST_METHOD from server
not ok 15 - CONTENT_LENGTH overflow ssite_t

Cheers,
Randall