The following URL might provide an idea what is going wrong here: http://www.coneural.org/florian/papers/04_byteserving.php
No guarantees that that's the reason but there seems to be a high probability because I've found a number of references to similar problems on the net. But that still doesn't point out what exactly needs to be done. But since I'm here at at the ApacheCon I've taken the opportunity to ask a couple of Tomcat users if they have additional ideas. The told me the following: - In the case of IE it is known that the Browser together with the Acrobat plug-in can make multiple requests against the server which can produce issues. - Use wget to retrieve a PDF from the same server to check that the PDF really arrives intact. If it works that way it indicates this verifies that there are multiple requests made by the browser that cause the problems. See byte serving above. See also some bugs that an HTTPD developer pointed me to. You can add follow-up information there if you have anything. But it's HTTPD, not Tomcat. https://issues.apache.org/bugzilla/buglist.cgi?query_format=advanced&short_desc_type=allwordssubstr&short_desc=pdf&product=Apache+httpd-2&long_desc_type=substring&long_desc=&bug_file_loc_type=allwordssubstr&bug_file_loc=&keywords_type=allwords&keywords=&bug_status=UNCONFIRMED&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&bug_status=NEEDINFO&bug_status=RESOLVED&bug_status=VERIFIED&bug_status=CLOSED&resolution=FIXED&resolution=WONTFIX&resolution=LATER&resolution=REMIND&resolution=WORKSFORME&resolution=MOVED&resolution=---&emailassigned_to1=1&emailtype1=substring&email1=&emailassigned_to2=1&emailreporter2=1&emailcc2=1&emailtype2=substring&email2=&bugidtype=include&bug_id=&votes=&chfieldfrom=&chfieldto=Now&chfieldvalue=&cmdtype=doit&order=Importance&field0-0-0=noop&type0-0-0=noop&value0-0-0= - Check the access log to see if what kind of requests are made against the webserver. - If this doesn't help you towards a solution, the Tomcat user list is certainly a good place to get more help. Please make sure that you mention version numbers for all components involved (Tomcat, Browser, Acrobat etc.). They told me that some bugs have been fixed in some versions of Tomcat that could relate to this problem. So it could be that a release change might help. Again no guarantees. If you find a solution elsewhere please lets us know about it in case anyone else runs into a similar problem in the future. On 08.04.2008 14:55:10 Woodhouse, Graeme wrote: > Oh sorry, forgot to add that info to the last email implicitly. > > I have tested adding a step of first streaming the file to disk and then > serving the file as a stream and that encounters the same problem. Giving a > direct link to the file on the server works fine but as I mentioned we don't > want to do that. > > Its quite difficult to make anything out of the file that is downloaded as > its in a binary format. I can tell that there isn't a set limit for size, > file seems to randomly fail between 25-30page+ mark. (Each of these pages > contains variable size images of around 1700x2000). > > Graeme Woodhouse > Software Engineer > ProQuest > > Direct-Line: +44 (0) 1223 271 264 > Fax: +44 (0) 1223 215 513 > > > -----Original Message----- > From: Manuel Mall [mailto:[EMAIL PROTECTED] > Sent: 08 April 2008 13:20 > To: 'fop-users@xmlgraphics.apache.org' > Subject: RE: Large PDF's files fail to stream correctly > > > > > -----Original Message----- > > From: Woodhouse, Graeme [mailto:[EMAIL PROTECTED] > > Sent: Tuesday, 8 April 2008 8:03 PM > > To: 'fop-users@xmlgraphics.apache.org' > > Subject: RE: Large PDF's files fail to stream correctly > > > > >Have you tried connecting to Tomcat directly and not via Apache to > > >confirm that the problem is related to the streaming via AJP? > > > > I've tried it now on a stand alone tomcat instance and am having the same > > trouble. It makes me even more confused that this isn't a more publicised > > problem on the internet. I would have thought many people dynamically > > generate large PDF's through tomcat (although writing that it does sound > > less likely with all those stipulations). > > > > So, it seems like its tomcat's problem more than the connection between > > apache and tomcat. > > I would still do the other suggested tests. Save the PDF as file and serve > that through tomcat and also capture the data at the client end and compare. > Jeremias Maerki --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
