Under what circumstances does Firefox issue a request requiring a 206 code response
I am having difficulty with images not downloading properly. I believe the problem is with the host but need to understand what is going on. For problematic images I see Firefox requesting an image and then later requesting a partial image. See below. Can anyone confirm that this second request is due to the server not fulfilling the first request properly.
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:14.0) Gecko/20100101 Firefox/14.0.1 FirePHP/0.7.1 Accept: image/png,image/*;q=0.8,*/*;q=0.5 Accept-Language: en-us,en;q=0.5 Accept-Encoding: gzip, deflate Connection: keep-alive x-insight: activate Range: bytes=12257- If-Range: Mon, 15 Oct 2012 20:54:09 GMT HTTP/1.1 206 Partial Content Date: Sun, 28 Oct 2012 12:56:00 GMT Server: Apache Last-Modified: Mon, 15 Oct 2012 20:54:09 GMT Accept-Ranges: bytes Content-Length: 5854004 X-Powered-By: PleskLin Content-Range: bytes 12257-5866260/5866261 Keep-Alive: timeout=3, max=199 Connection: Keep-Alive Content-Type: image/jpeg
Изменено
Все ответы (5)
That request show a byte range with only a starting value: Range: bytes=12257-
So it seems that Firefox only got the first 12256 bytes of that 5.8MB image for some reason.
Do you have request headers for that first part?
Do you have security software that may be scanning images and has a problem with such large images?
Thanks cor-el. I dont think there is any program on my PC that is interfering. You see, when I upload the test to another server it works ok. So it seems to be the host that is at fault. But I wanted to be sure - e.g. it was not Firefox giving up cos the host was slower or something. So I wanted to understand why firefox would send out this request. Then I can go to the hosting company with proof that they are at fault. The full set of headers are here...
https://skydrive.live.com/redir?resid=9F8DE44BC9AC7FD6!110&authkey=!AA0XFU0wS2DE0lE
and result i see is here... https://skydrive.live.com/redir?resid=9F8DE44BC9AC7FD6!107&authkey=!AAXurFZleJE5okQ
Изменено
I've checked that page and sometimes get the same "Image corrupt or truncated errors" in the Web Console, so I assume that it is a problem with that host that aborts the request.
You do load 9 images each over 4MB on that test page that are scaled down to 100x100 and this is not really recommended.
The page takes a long time to load because of this.
Thanks for the reply cor-el. Yes it is a lot of data - this is just a test. By doing this I can get more failures than normal which makes it easier to benchmark. I should point out that when I repeat the test on another host company that I have access to, I do not get any failures. Even when I run 3 tests in parallel. (It takes a very long time to complete then - but all images appear eventually).
Best would be to contact the hosting company and ask them if they have configured their servers to respond in this way, possibly if their servers are (too) busy with serving content to other visitors.