Question

We have a system where a client makes an HTTP GET request, the system does some processing on the backend, zips the results, and sends it to the client. Since the processing can take some time, we send this as a ZipOutputStream wrapping the response.getOutputStream().

However, when we have an exceptionally small amount of data in the first ZipEntry, and the second entry takes a long time, the browser the client is using times out. We’ve tried flushing the stream buffer, but no response seems to be sent to the browser until at least 1000 bytes have been written to the stream. Oddly, once the first 1000 bytes have been sent, subsequent flushes seem to work fine.

I tried stripping down the code to bare-bones to give an example:

protected void doGet(HttpServletRequest request,
        HttpServletResponse response) throws ServletException, IOException {
    try {
        ZipOutputStream _zos = new ZipOutputStream( response.getOutputStream());
        ZipEntry _ze = null;
        long startTime = System.currentTimeMillis();
        long _lByteCount = 0;

        response.setContentType("application/zip");

        while (_lByteCount < 2000) {
            _ze = new ZipEntry("foo");
            _zos.putNextEntry( _ze );

            //writes 100 bytes and then waits 10 seconds
            _lByteCount += StreamWriter.write( 
                    new ByteArrayInputStream(DataGenerator.getOutput().toByteArray()),
                    _zos );
            System.out.println("Zip: " + _lByteCount + " Time: " + ((System.currentTimeMillis() - startTime) / 1000));

            //trying to flush
            _zos.finish();
            _zos.flush();
            response.flushBuffer();
            response.getOutputStream().flush();
        }
    } catch (Throwable e) {
        e.printStackTrace();
    }
}

I set my browser timeout to be about 20 seconds for easy reproduction. Despite writing the 100 bytes a couple of times, nothing is sent to the browser and the browser times out. If I expand the browser timeout, nothing gets sent until 1000 bytes have been written and then the browser pops up the “Would you like to save…” dialog. Again, after the initial 1000 bytes, each addition 100 bytes sends fine, rather than buffering to 1000 byte chunks.

If I set the max byte count in the while condition to 200 or so, it works fine, sending only 200 bytes.

What can I do to force the servlet to send back really small initial amounts of data?

Answer 1

It turns out there is a limit on the underlying Apache/Windows IP stack that buffers data from a stream in an attempt to be efficient. Since most people have the problem of too much data, not the problem of too little data, this is right most of the time. What we ended up doing was requiring the user to request enough data that we’d hit the 1000 byte limit before timing out. Sorry for taking so long to answer the question.