-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Miguel,
On 8/22/12 4:26 PM, Miguel González Castaños wrote: > On 22/08/2012 18:06, Christopher Schultz wrote: >> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 >> >> Miguel, >> >> On 8/21/12 10:24 AM, Miguel Gonzalez wrote: >>> Our webapp worked fine under Tomcat 5.5. We decided to migrate >>> to Tomcat 7 to get better performance and get support from the >>> community in case something goes wrong as suggested in this >>> mailing list. >>> >>> I installed tomcat 7.0.29 from the Apache website. I had >>> issues downloading PDFs from IE, but not from Firefox or >>> Chrome. >>> >>> I googled a little bit and some people suggested to downgrade >>> to 7.0.26. But still having issues with it under IE while it >>> works under Firefox and Chrome. >>> >>> Any ideas? >> http://www.catb.org/esr/faqs/smart-questions.html >> >> Your question is "I had an app that was working and then I >> upgraded and it doesn't work. What might be the problem?". > Not exactly, I said problems downloading PDFs in IE. It reported > an error when tried to download like it couldn't finish off the > downloading (it takes forever). >> >> Well, the problem might be that the power isn't connected to >> your server. But you haven't provided any information that >> suggests what "had issues downloading PDFs from IE" means. What >> issues? What did you try? What happened when you did that? What >> have you attempted to do so far to "fix" the "issue" and what >> have you discovered? > In the end downgrading to 7.0.26 has solved the issue. The person > that reported that he had also issues with 7.0.26 and IE8 > recognized it might be his install since it happened just with a > particular file, not all PDF files. I couldn't reproduced the issue > with IE8 :( Obviously reproducability is a key ingredient to proper analysis. If you and he are willing to do some targeted testing, you could instrument the client and see what the difference is -- likely in the headers -- in the response coming from the server. It might come down to a combination of Pragma, eTag, and other headers coming from Tomcat (or an intermediate proxy). There have been reports of weirdness with Adobe Acrobat Reader and "Range" headers recently: it appears that the client is only partially spec-compliant because the server is not required to return the exact range(s) the client requested. - -chris -----BEGIN PGP SIGNATURE----- Version: GnuPG/MacGPG2 v2.0.17 (Darwin) Comment: GPGTools - http://gpgtools.org Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAlA2YZMACgkQ9CaO5/Lv0PABBQCfYKYDydwXBKRa9Y/R/BJhIs0H L+0AnjZeVD4nHgvTg1SlSh2dOK9c7xcO =dip4 -----END PGP SIGNATURE----- --------------------------------------------------------------------- To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
