Firefox 技術支援討論區

About:telemetry shows an ID value in General Data. Is this the basis for firefox calculating the number of active users? Because the ID value is different for my across different versions of Firefox on the same computer.

In a reddit thread, someone has made this claim:

" MAUs are industry standard KPIs. Usernames, emails, user ids are tied to guids or uuids for the purpose of ensuring that you track unique users. Anyone who has used software like Segment understands this.

   Mozilla needs to know its platform usage. This comes from device identifiers and os data, and is used in tandem with user data to track the statistics of it’s MAUs. Any software firm or website runner will be able to confirm this.

So when you use Mozilla and create a username, they tie that data to the browser and user info they get from you using their platform. Because that’s what you’re doing: using their software."

I do not believe this statement could be correct, because it seems contradictory to https://wiki.mozilla.org/Security/Anti_tracking_policy

I have spent an hour researching Firefox articles on telemetry, privacy policy and I still do not know what is the basis of the active users statistic. I have read that my IP address is logged in addition to the telemetry data visible in the about:telemetry page, but this does not mean that my log in with Firefox accounts is used as part of telemetry data or in tracking my usage of Firefox.

I have attached troubleshooting data to see what is included.

I've been using firefox for years, mostly on mac and windows. I turned my mac on this morning and when I tried to go fullscreen, the browser froze and eventually closed itself a minute later. I've tried reinstalling, using different versions of firefox, but nothing seems to fix it. Firefox Nightly doesn't freeze, but it crashes instantly when fullscreening. Any idea why any of this is happening? It was working perfectly fine yesterday, I don't believe it was an update because I manually updated it after the bug appeared to see if it would fix it. Thanks!

MacOS: 12.0 Monterey Firefox Version: 98.0b7

Hi.

- Firefox is set to "always ask to you where to save files" and this works fine I when i choose to save a file.

- Multiple file types have been set to "always ask" so it can give me the ability to "Open with" (an application) a document rather than downloading it. However, I expect the (temporary) downloaded document to be (opened) in the temp folder, but it's actually being saved in my Downloads folder.

That's not comfortable, but i end up with a massive lit of files in the download folders which were never intended to be saved in the first place, but just opened. That gets mixed up with other stuff in downloads folder.

How can i force the behavior "Open With" to save the document in firefox temp folder rather Downloads ? Thanks for your help. Regards. Flo

How to set the icon to be what profile is selected around the address bar of the website? Similar to mozilla account only for profile. I don't mean about: profile. Only that when I browse the website, e.g. https://www.mozilla.org/en-US/, somewhere next to the address, I see information about the currently used profile. Because now I can only guess because the tabs and the history because the profiles are different.

ZoogVPN support wrote: «modern browsers (Chrome, Firefox) don't support authentication with SOCKS5, they only support it with HTTP proxies». Is this really true? Mozilla Firefox does not prompt or save username and password when trying to connect to a private SOCKS v5, and the setting "Connection Settings" / "Configure Proxy Access to the Internet" / "Do not prompt for authentication if password is saved" cannot be used to use logins and passwords from SOCKS v5?

hi, not sure where to start this issue as its been building up over the last 1-2 months more and more. i had brushed it off for a bit but its become quite apparently an issue on Firefox and nothing else.

the main thing is that when i click into different tabs, certain pages turn blank (for example, when i away from Youtube i got back and the page is just white) until i click again. but this also occurs on other pages in a way where anything not present is loaded but scrolling down its blank. this happens a lot with tumblr or twitter, and often times i need to completely refresh to fix it. it often lags too when this happens, the pages being slow to respond. i have had other pages where you need to click or scroll to load feeds (soundcloud for example) where they just wont load the content after i scroll and it remains blank.

i thought it was specific sites and not the browser but its apparent its just firefox. chrome works fine and so does Edge. i have looked at solutions here related to issues with loading but none seem to fit the bill for my issue and arent relevant/dont seem to change things.

I use eBay on a regular basis, and always start with ebay.ie because it is in English and has euros as the default currency. However, at times, links within eBay will point to ebay.co.uk instead of ebay.ie, and that will result in browsing outside of the EU, under different cookie legislation and different governing agreements. When I notice such a link, I can replace the .co.uk with .ie, and the page will work perfectly. However, I would love to have the ability to do this automatically.

I understand that the server may be the same but what matters, when it comes to the settings that eBay applies and to the user agreement applied to me, is whether the request has a .co.uk or a .ie in it. Is there a way to automate this?

I am developing a JavaScript CSS HTML Canvas application using VS Code and Firefox. As the code renders graphics to the canvas - the size of the canvas is critical. Identifying chages in screen size works perfectly for normal resizing events. However - going into fullscreen mode does not work. I have tried a number of different event listeners:

document.addEventListener('fullscreenchange', setCanvasDimensions, false);

document.addEventListener('mozfullscreenchange', setCanvasDimensions, false);
document.addEventListener('MSFullscreenChange', setCanvasDimensions, false);
document.addEventListener('webkitfullscreenchange', setCanvasDimensions, false); 

None of these events fire when the screen goes into fullscreen mode. My apologies if this has already been addressed elsewhere or if this is a total newbie problem - I have searched for a solution although none seem to work. Any help with this would be appreciated. Stay safe.

why do these javascript commands not work: all I get is a new blank tab. Javascript is enabled in about:config.

javascript:WebForm_DoPostBackWithOptions(new WebForm_PostBackOptions("ctl00$oSaveBar$btnSave", "", true, "", "", false, true)) 

javascript:__doPostBack('ctl00$cphEdit$rptDocs$ctl00$lnkPdf',)

I think this went missing in a recent update. Chrome seems to have hijacked it. I reset the email app in windows settings and mailto showed back up in ff, but it no longer shows gmail as the action.

Sometimes when loading a page, and the page is slow to respond, what is expected is seeing a notice of error, allowing me to refresh the page. However, sometimes the strange behaviours happen:

• The refresh button is gray. I cannot refresh. I can only to re-type in my URL bar to force reload the page; or
• The URL bar keeps blank or remains the former page. For example, I click the link on page A to page B, and page B is too slow, then I found that the URL keeps the URL of page A. I have to re-click the link of page B on page A; if I opens page B on a new page, then the URL keeps blank finally, and I have to re-click the link of page B as well.

The operation system is Windows 11. It also happens on Windows 10. It happens on Firefox and Firefox Developer Edition. The issue does not exist in Edge browers (in Edge, it displays connection failure and allows me to refresh).

The issue has been existing for years.

See the previous archived discussion: https://support.mozilla.org/zh-CN/questions/1354935

Hi...

Is there a way to edit bookmarks sidebar right-click menu. I want to change position of "Delete folder" and "Delete bookmark" to bottom of right-click menu.

Thank you

The issue I am experiencing is this - whenever I refresh a tab within my browser while developer tools is open on that tab, the DOM tree disappears completely. I am still able to click "inspect" and hover over elements, and I still get the visual feedback of the styles being applied to that element. Additionally, the CSS styles still display in the right-hand side of developer tools when I inspect an element. The only issue I'm seeing within the "Inspector" tab within developer tools is the missing DOM tree.

The other tabs within developer tools (Console, Debugger, Network, Performance, etc...) seem to be working fine (as far as I can tell). Navigating to another page within this tab does not solve the issue.

If I open a new tab within the same browser window, I am able to see the DOM tree again, but only until the page refreshes in which case the error occurs again. I have experienced this issue intermittently for a couple of years now, but this is the first time where it has been persistent despite re-installing browser, restarting computer, etc...

Please let me know if there is any further information that I can provide that might help solve this issue.

Thanks very much for any help that anybody can provide.

For example: I visit https://raw.githubusercontent.com/vvb2060/Magisk/b03fdcb14dc0e7ec88e98ac3dd769d96ab01c1e2/app/src/main/res/raw/changelog.md It open normally and I can directly see what inside of it. The "content-type" of this http request is "text/plain".

But if I visit https://cdn.jsdelivr.net/gh/vvb2060/Magisk@b03fdcb14dc0e7ec88e98ac3dd769d96ab01c1e2/app/src/main/res/raw/changelog.md Firefox will download it automatically, instead of directly open it in browser. The "content-type" of this http request is "text/markdown".

This is not the problem of "What should Firefox do with other files?" setting...

I make a test, I create a file "test.md" in the Nginx website folder, and I add "default_type text/markdown" to conf to force its content-type, pic here: https://vip2.loli.io/2022/05/12/v39nT8wu6yIlXeH.png

When I open it in Edge Chromium, it will shows what inside of it: https://vip1.loli.io/2022/05/12/EGFwgHPtNjqshyl.png

But Firefox... https://vip1.loli.io/2022/05/12/vuz8YBQAlD2ZCGk.png ( Here I turn on "Ask whether to open or save files")

I am using Firefox Dev 101.0b5 and also test it with Firefox Stable 100.0, the same.

Hello,

It got really hard to debug with Firefox Developer Edition devtools when inspector tab keeps disappearing while debugger point is hit.

Using HTMLElement link from console doesn't return Inspector

Pressing F12 and F12 again helps, but then again have to reach the break point which is frustrating.

Firefox 101.0b5 (64-bit) Windows 10 Pro 21H2 Build 19044.1645 CPU - i5-1145g7 RAM - 16gb

Any way to solve this apart reinstalling Firefox ? As well why the same bug can stay 8 years? https://support.mozilla.org/en-US/questions/980988

duplicate of /questions/1376840 thread.

found a bug with firefox 101.0b5 website https://www.express.co.uk/ it does not display the comments for an article. Works with Microsoft Edge

I'm not able to get hardware accelerated video decode working on Firefox developer 101.0b7 on Arch linux X11 with VA-API. This is verified by a lack of Video activity in intel-gpu-top

There is also no obvious error message. On youtube videos I sometimes see

[2022-05-16T17:36:32Z ERROR mp4parse] Found 2 nul bytes in "\u{0}\u{0}" [2022-05-16T17:36:32Z ERROR mp4parse] Found 2 nul bytes in "\u{0}\u{0}" [2022-05-16T17:36:32Z ERROR mp4parse] Found 2 nul bytes in "\u{0}\u{0}" [2022-05-16T17:36:32Z ERROR mp4parse] Found 2 nul bytes in "\u{0}\u{0}"

but on Big Buck Bunny from https://test-videos.co.uk/vids/bigbuckbunny/mp4/h264/1080/Big_Buck_Bunny_1080_10s_1MB.mp4 I don't get that output and everything looks normal:

libva info: VA-API version 1.14.0 libva info: Trying to open /usr/lib/dri/iHD_drv_video.so libva info: Found init function __vaDriverInit_1_14 libva info: va_openDriver() returns 0

but the video decode in intel-gpu-top shows 0%.

VA-API is confirmed to be working in mpv, if I play that same big buck bunny video in mpv I see 10-20% video decode usage.

vainfo shows

vainfo: VA-API version: 1.14 (libva 2.14.0) vainfo: Driver version: Intel iHD driver for Intel(R) Gen Graphics - 22.4.1 () vainfo: Supported profile and entrypoints

     VAProfileNone                   :	VAEntrypointVideoProc
     VAProfileNone                   :	VAEntrypointStats
     VAProfileMPEG2Simple            :	VAEntrypointVLD
     VAProfileMPEG2Simple            :	VAEntrypointEncSlice
     VAProfileMPEG2Main              :	VAEntrypointVLD
     VAProfileMPEG2Main              :	VAEntrypointEncSlice
     VAProfileH264Main               :	VAEntrypointVLD
     VAProfileH264Main               :	VAEntrypointEncSlice
     VAProfileH264Main               :	VAEntrypointFEI
     VAProfileH264Main               :	VAEntrypointEncSliceLP
     VAProfileH264High               :	VAEntrypointVLD
     VAProfileH264High               :	VAEntrypointEncSlice
     VAProfileH264High               :	VAEntrypointFEI
     VAProfileH264High               :	VAEntrypointEncSliceLP
     VAProfileVC1Simple              :	VAEntrypointVLD
     VAProfileVC1Main                :	VAEntrypointVLD
     VAProfileVC1Advanced            :	VAEntrypointVLD
     VAProfileJPEGBaseline           :	VAEntrypointVLD
     VAProfileJPEGBaseline           :	VAEntrypointEncPicture
     VAProfileH264ConstrainedBaseline:	VAEntrypointVLD
     VAProfileH264ConstrainedBaseline:	VAEntrypointEncSlice
     VAProfileH264ConstrainedBaseline:	VAEntrypointFEI
     VAProfileH264ConstrainedBaseline:	VAEntrypointEncSliceLP
     VAProfileVP8Version0_3          :	VAEntrypointVLD
     VAProfileVP8Version0_3          :	VAEntrypointEncSlice
     VAProfileHEVCMain               :	VAEntrypointVLD
     VAProfileHEVCMain               :	VAEntrypointEncSlice
     VAProfileHEVCMain               :	VAEntrypointFEI
     VAProfileHEVCMain10             :	VAEntrypointVLD
     VAProfileHEVCMain10             :	VAEntrypointEncSlice
     VAProfileVP9Profile0            :	VAEntrypointVLD
     VAProfileVP9Profile2            :	VAEntrypointVLD

As far as I know all flags are set properly (according to https://wiki.archlinux.org/title/firefox#Hardware_video_acceleration) and I have set MOZ_DISABLE_RDD_SANDBOX=1

Good day.

My use case is the following. I use Firefox Developer Edition for developing websites. Usually the development happens on a certain part of the page. Most of the times it is not on the top of the page, so you have to scroll down to it.

Sometimes when the page is refreshed a jump to the top happens. So then I need to scroll down again. It takes a lot of time and concentration away if it happens dozens of times a day.

It often happens after I did some changes to the files on the page. It usually does not happen when I refresh the page a couple of times (when I purpously want to reproduce this problem).

I did try running Firefox in Safe mode, but it didn't help.

What can be done to fix this problem on my side? Is this a bug in Firefox itself?

Hello,

I am trying to download Firefox source code. I tried the simplest option offered to me : https://developer.mozilla.org/en-US/docs/Mozilla/Developer_guide/Using_the_VM But the link to the Virtual Machine does not work. Do you know if the site will be back soon or will I have to search for another website offering this VM ?

Sorry for my bad english. I hope my question was clear

Clement