httpd-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Alex Moon <A.M...@mdx.ac.uk>
Subject RE: apache 2.0.39 under NT large pdf file failure
Date Thu, 08 Aug 2002 13:01:34 GMT
Owen - Thanks for this its a start point - i'll try out all of it but a 
download by save target does give a correct file that opens fine but:

Save target files open OK which is a plus but:
Original file size 35833
Inetexplorer 5 and 6 downloads is 35833
Netscape 4.7 downloads is 36783

Yet netscape seems to nearly always work, its IE that seems to 
have the greatest problems.  Netscape has always been a bit odd 
like this though and i blame the browser.  

I will try investigating your ideas about a header causing the 
problem but not sure what you mean by hit by the command line.

Here are a few more pieces to the puzzle:

have had the occasional error in Netscape:-

 unknown staus reply from server:0!

Then if you try and load the page again it loads.
Generally though, pdf files simply don't work either blank or run cpu 
usage up to 100% (Acrobat Process) or give file corruption error.

The server is an intranet so can't make it available but i could set 
up a temporary pc with the same problem.

Alex


On 8 Aug 02, at 11:59, Boyle Owen wrote:

> I have heard of problems like this before... I am not entirely certain what causes it
but I believe it is to do with the way IE handles PDF files. First of all, it fires up acrobat
in the browser then, as the file starts to come in, it tries to open it as soon as it has
enough data to display 
anything (rather like the way a browser will "unroll" a jpg image as it comes in). Why this
doesn't seem to work with apache 2, I've no idea but I could propose the following experiments
to try to understand the problem:
> 
> - If you hit the file on a 1.3 server, does the PDF start to display immediately (indicating
that display starts before download is complete) OR do you have to wait until download is
really complete before anything is displayed?
> 
> - On the 2.0 server: Instead of just clicking on the link to the PDF (which will automatically
fires up acrobat), right-click then "Save Target As..." and save the file to the Desktop.
Then start acrobat (i.e. separately from IE) and try to load the file - does it display correctly?
> 	
> 	If it doesn't, check the file properties (byte-size etc. - or even calculate a checksum)
- do they match what is on the server? Is there any evidence of corruption?
> 
> 	If it does open cleanly, then it cannot be corruption during download.
> 
> - If you download with other browsers (Netscape, Opera) what happens?
> 
> If there is some problem about acrobat trying to read the file as it arrives, it can't
be doing this based on the extension (or it would do it with the 1.3 server as well). So it
must be doing so based on a header from the server.
> 
> Can you hit the 2.0 server and the 1.3 server from the command line so you can look at
the HTTP headers that precede the file? Look for any differences and particularly something
like:
> 
> StartDisplayDuringDownload: Yes
> 
> If you found it, you can disable the header with the "Header" directive...
> 
> ...It's a long shot, but it might just work.
> 
> Rgds,
> 
> Owen Boyle
> 
> PS If you happen to have the 2.0 server on the internet, send me the URL and I'll have
a look at 
the headers it sends..
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@httpd.apache.org
> For additional commands, e-mail: users-help@httpd.apache.org
> 



---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@httpd.apache.org
For additional commands, e-mail: users-help@httpd.apache.org


Mime
View raw message