Mime Type issue in Internet Explorer

[Richard Norris]

Hello,
We recently moved from a Windows install of 3.4.4 Bugzilla to 4.4.4. We have run into an issue when a user uploads a Word document in docx format. When another user wants to view the docx document Bugzilla shows an error Attachment is not viewable in your browser because its MIME type (application/vnd.openxmlformats-officedocument.wordprocessingml.document) is not one that your browser is able to display . If you click on the download the attachment tab then you get a new broswer screen with giberish. This does not happen in Chrome. Chrome does still say the attachment is not viewable but it allows you to click download the attachment and it successfully downloads. Also I have gone back to our old Bugzilla 3.4.4 server and in the old enviroment you can view that same document or save it without issue.

Let me also add that I verified that application/vnd.openxmlformats-officedocument.wordprocessingml.document is in Apaches mime types configuration file so I do not know why it is unable to open that type of attachment.

Also if I turn off Allow attachment display in parameters in the new 4.4.4 enviroment then I am able to download the attachment in internet explorer. But since most of our attachments are not word documents then turning off the ability to view attachments in the browser really is not an option. Does anyone know if there are other configuration files that control mime type for Bugzilla? So far all of my research has pointed me to the mime type file in apache program files.

[Thorsten Schöning]

Guten Tag Richard Norris,
am Montag, 2. März 2015 um 21:32 schrieben Sie:

> If you click on
> the download the attachment tab then you get a new broswer screen
> with giberish. This does not happen in Chrome. Chrome does still say
> the attachment is not viewable but it allows you to click download
> the attachment and it successfully downloads.

From my experience Chrome implements a workaround to recognize the new
docx formats on it's own, regardless if the browser sends
application/zip or such, which in most cases is the problem.

> Also I have gone back
> to our old Bugzilla 3.4.4 server and in the old enviroment you can
> view that same document or save it without issue.

Your Chrome renders docx in your old Bugzilla? In any case, use the
dev tools of your IE and Chrome with your old Bugzilla and compare the
results specially regarding Content-type headers with your new one.

> Let me also add that I verified that
> application/vnd.openxmlformats-officedocument.wordprocessingml.document
> is in Apaches mime types configuration file so I do not know why it
> is unable to open that type of attachment.

That doesn't say much, this configuration is only used unless an
application overrides it. Check with your browser dev tools what
Bugzilla actually sends.

Your problem has been reported on the list before, but without any
result or feedback. You may want to search the list for hints, though.

Mit freundlichen Grüßen,

Thorsten Schöning

--
Thorsten Schöning E-Mail: Thorsten....@AM-SoFT.de
AM-SoFT IT-Systeme http://www.AM-SoFT.de/

Telefon...........05151- 9468- 55
Fax...............05151- 9468- 88
Mobil..............0178-8 9468- 04

AM-SoFT GmbH IT-Systeme, Brandenburger Str. 7c, 31789 Hameln
AG Hannover HRB 207 694 - Geschäftsführer: Andreas Muchow

[Richard Norris]

Thorsten,
Thank you for the response. I have tried what you suggested and it appears that when users upload docx files and they choose auto-detect it is choosing plain text as the content type. If I choose to select from list and choose binary file application/octlet-stream then it allows me to open the document. Can you help me understand how to fix the issue about auto-detect choosing the wrong content type? Is there a way to say docx = application/octet-stream. Thanks again for your help. -Chip

[Thorsten Schöning]

Guten Tag Richard Norris,
am Dienstag, 3. März 2015 um 16:33 schrieben Sie:

> Can you help me understand how to fix the issue about auto-detect
> choosing the wrong content type? Is there a way to say docx =
> application/octet-stream.

No, sorry, but if you want to have a further look yourself, start
around Bugzilla::Attachment::_check_content_type to see where the
problems is introduced, especially if Bugzilla autodetects text/plain
right from the start or only after it got octet-stream or such. The
information will help the devs as well, if you decide to file a bug
later.

http://www.bugzilla.org/developers/reporting_bugs.html

[Hildebrand, Jeff]

We had the same problems with IE 9. Both Firefox and Chrome handled it by downloading a copy of the file then opening it and the data was correct. But with IE 9 it seemed to have an issue with trying to access the data directly and not recognizing the proper type. When we tested with IE 11, it worked correctly.

Given there are other problems with IE 9, at this time we are telling our users to use Firefox and Chrome until our IT can install IE 11 for everyone.

So check your version of IE and see if 11 fixes it for you too.

Jeffery Hildebrand | Mixed Signal EDA Engineer
Voice Products - CPG
Office: +1 858-675-3469 | Cell: +1 (952) 210-8926
www.microsemi.com