Closed Bug 1506173 Opened 10 months ago Closed 10 months ago

Quit message confusing - implies all tabs will be lost when exiting browser

Categories

(Firefox :: Session Restore, defect)

defect
Not set

Tracking

()

VERIFIED FIXED
Firefox 65
Tracking Status
relnote-firefox --- 65+
firefox-esr60 --- unaffected
firefox64 --- unaffected
firefox65 --- verified

People

(Reporter: jesup, Assigned: Gijs)

References

(Depends on 1 open bug, )

Details

(Keywords: regression)

Attachments

(1 file, 1 obsolete file)

The new state of putting up a "Do you want to close 485 tabs?" on Quit/Exit is very confusing and alarming, especially for those who have "Restore previous session" (tabs) set.  It's the same message for closing a window (which can lose tabs).  

Note that on Windows and Linux, closing the last window exits the browser, and if it wouldn't (another window still open), you get the warning (since you will lose tabs if you continue).  On Mac, we typically keep running even if the last window is closed, and so you *do* lose tabs if you close the last window...

Due to this difference, several options present themselves:
1) go back to the old behavior on Windows and Linux, and keep the new behavior on Mac
1a) Same, but only hide the warning on Windows/Linux if Restore Session is set (though you can Restore Previous Session even if you do shut down with tabs open).
1b) same as 1 or 1a, with added option on Mac to instead Exit the browser and not lose tabs.
2) Change the wording to be clear on Windows/Linux that we're closing the browser - "Exit the browser now? (N tabs will be closed)" (if Restore isn't set), or "Exit the browser now? (N tabs will be restored on next start of Firefox)" (or some such wording) if Restore is set.
Blocks: 550559
No longer depends on: 550559
Just for the record, just like individual tabs and the entire session, you can recover closed windows using the history menu in any remaining windows (or the ctrl-shift-n shortcut).

So,

(In reply to Randell Jesup [:jesup] from comment #0)
> The new state of putting up a "Do you want to close 485 tabs?" on Quit/Exit
> is very confusing and alarming, especially for those who have "Restore
> previous session" (tabs) set.  It's the same message for closing a window
> (which can lose tabs).  

I disagree that this is true. You don't "lose" anything, in either case.

> Note that on Windows and Linux, closing the last window exits the browser,
> and if it wouldn't (another window still open), you get the warning (since
> you will lose tabs if you continue).  On Mac, we typically keep running even
> if the last window is closed, and so you *do* lose tabs if you close the
> last window...

And as a result, I strongly feel we shouldn't make any platform-specific changes here.

This whole rigmarole started because bug 1438499 was confused that we surface a dialog that closes the last window, but don't use the same dialog for quitting (with automatic session restore disabled). When that was fixed, people complained that now there was no way to get a warning dialog for quit, if you had session restore enabled. Previously, there had been an about:config pref for that, with a completely *different* dialog - because 2 dialogs for quitting wouldn't make sense, I had removed the other dialog that users couldn't enable outside of about:config anyway. Then we fixed warning in the session restore case in bug 550559 by using the same dialog everywhere (and being, arguably, much more consistent), and now people complain because they want the opposite again. The winning move is not to play and leave the dialogs in whatever state they're in, clearly.

If you want a warning for closing windows and not for quitting, there's still an about:config pref to get what you want - set browser.warnOnQuit to false.

(In reply to Randell Jesup [:jesup] from comment #0)
> 2) Change the wording to be clear on Windows/Linux that we're closing the
> browser - "Exit the browser now? (N tabs will be closed)" (if Restore isn't
> set), or "Exit the browser now? (N tabs will be restored on next start of
> Firefox)" (or some such wording) if Restore is set.

We can change the warning to mention whether the tabs will be automatically restored or not, and to be explicit about whether we're quitting/exiting or not. But I don't think there should be any platform differences - there's no reason for that.

Markus/Amy, can you clarify whether you think that if people have session restore enabled we should mention Firefox will restore tabs, right in the dialog (in the 'quit' case)? Also, should we be explicit in the dialog about whether Firefox will quit, or "just" close a single window? If so, can you and/or Meridel suggest wordings for the 4-8 different strings we'd need to present (as opposed to 2 right now - 1 for a single window, 1 for multiple windows)?
Flags: needinfo?(mjaritz)
Flags: needinfo?(amlee)
I'd argue that closing 485 tabs *is* losing something. You don't know which state they had. Some tabs may have transient state in forms, in web page state that will not be reproduced on reload, etc.. Some users might have prefs that mean a reload does drop things. For example, I have intentionally set the prefs so that cookies are cleared on shutdown. But I don't want cookies to clear and re-login everywhere just because I accidentally hit Ctrl-Q (which I never want) instead of Ctrl-W (which I do all the time to close tabs).

I think the current state is fine.

If people want a slightly (!) different message depending on whether the window is closed or all windows are closed or the entire browser instance is closed, that would be fine. The differences should be small in wording, though, like:
* "Do you want to close this window with 5 tabs?"
* "Do you want to close all windows with 485 tabs?"
* "Do you want to exit the browser and close all windows with 485 tabs?"

But we definitely do need these warnings. The lack of it was a very annoying bug and a very, very common user complaint (bug 52821) and a very obvious FAIL to not warn. I think most people *would* want to know before they close and reload 485 tabs.
While I find it annoying to my workflow (I have to explicitly Exit (or ctrl-shift-q) frequently because I run ASAN Nightly on one machine, and do occasional Exits on my "Main" browser because it's a local m-c build, not a Nightly with auto-update -- I can see Ben's point.  His wording isn't bad - it's much more explicit about what will happen, though something about Restore enabled would be good to add, if set.  

I'll note to Ben's argument that Exit and Restore is losing something and deserves a warning -- that would apply to applying an update as well, which we as for permission for - but we give no warning that state in tabs could be lost (we say we'll restore the user's tabs, but really it's Restore as in close and reload, just like exit).
(In reply to :Gijs (he/him) from comment #1)
> Just for the record, just like individual tabs and the entire session, you
> can recover closed windows using the history menu in any remaining windows
> (or the ctrl-shift-n shortcut).
> 
> So,
> 
> (In reply to Randell Jesup [:jesup] from comment #0)
> > The new state of putting up a "Do you want to close 485 tabs?" on Quit/Exit
> > is very confusing and alarming, especially for those who have "Restore
> > previous session" (tabs) set.  It's the same message for closing a window
> > (which can lose tabs).  
> 
> I disagree that this is true. You don't "lose" anything, in either case.
> 
> > Note that on Windows and Linux, closing the last window exits the browser,
> > and if it wouldn't (another window still open), you get the warning (since
> > you will lose tabs if you continue).  On Mac, we typically keep running even
> > if the last window is closed, and so you *do* lose tabs if you close the
> > last window...
> 
> And as a result, I strongly feel we shouldn't make any platform-specific
> changes here.
> 
> This whole rigmarole started because bug 1438499 was confused that we
> surface a dialog that closes the last window, but don't use the same dialog
> for quitting (with automatic session restore disabled). When that was fixed,
> people complained that now there was no way to get a warning dialog for
> quit, if you had session restore enabled. Previously, there had been an
> about:config pref for that, with a completely *different* dialog - because 2
> dialogs for quitting wouldn't make sense, I had removed the other dialog
> that users couldn't enable outside of about:config anyway. Then we fixed
> warning in the session restore case in bug 550559 by using the same dialog
> everywhere (and being, arguably, much more consistent), and now people
> complain because they want the opposite again. The winning move is not to
> play and leave the dialogs in whatever state they're in, clearly.
> 
> If you want a warning for closing windows and not for quitting, there's
> still an about:config pref to get what you want - set browser.warnOnQuit to
> false.
> 
> (In reply to Randell Jesup [:jesup] from comment #0)
> > 2) Change the wording to be clear on Windows/Linux that we're closing the
> > browser - "Exit the browser now? (N tabs will be closed)" (if Restore isn't
> > set), or "Exit the browser now? (N tabs will be restored on next start of
> > Firefox)" (or some such wording) if Restore is set.
> 
> We can change the warning to mention whether the tabs will be automatically
> restored or not, and to be explicit about whether we're quitting/exiting or
> not. But I don't think there should be any platform differences - there's no
> reason for that.
> 
> Markus/Amy, can you clarify whether you think that if people have session
> restore enabled we should mention Firefox will restore tabs, right in the
> dialog (in the 'quit' case)? Also, should we be explicit in the dialog about
> whether Firefox will quit, or "just" close a single window? If so, can you
> and/or Meridel suggest wordings for the 4-8 different strings we'd need to
> present (as opposed to 2 right now - 1 for a single window, 1 for multiple
> windows)?

I agree we should be more explicit in our wording for these warning messages. Especially if Restore session is enabled or not we should tell the user if their tabs will be restored when we open the browser again. NI Meridel for copy suggestions for warning dialogue.
Flags: needinfo?(amlee) → needinfo?(mwalkington)
I also found the new quit warning confusing and scary when I first saw it since it is normally associated with losing tabs unless the user makes a conscious effort to recover them later. There is a big difference between closing a single window and quitting the application, yet they both share the same dialog and the same setting (browser.warnOnQuit = true doesn't work if browser.tabs.warnOnClose = false). There is still no warning when quitting with a single tab in a single window despite significant data loss if the user has chosen to clear history on exit.

What most people seem to be calling for in the 4 duplicates of Bug 550559 and the 24 duplicates of Bug 502908 (12 of them by the same two people) is a dedicated quit confirmation dialog.
Duplicate of this bug: 1506069
Duplicate of this bug: 1506161
Hi Amy,

Can you help me understand the interactions that need copy? I see 4-8 strings are mentioned. 

Can you please define and describe each distinct interaction? This will help me develop the content.

Thanks!
Flags: needinfo?(mwalkington) → needinfo?(amlee)
If the "session restore" preference setting is set, and you think it's important enough to merit inclusion, maybe a note could be added:

* "Do you want to close this window with 5 tabs?"
* "Do you want to close all windows with 485 tabs? Your tabs can be reloaded after restart."
* "Do you want to exit the browser and close all windows with 485 tabs? Your tabs can be reloaded after restart."

(I intentionally didn't write "restored", because there may very well be state in tabs that we cannot restore.)
(FWIW, while the warning is important, brevity is key here, so personally, I would not include this note. The user will find out himself anyway, the first time he restarts, that we allow to reload tabs.)
(In reply to Ben Bucksch (:BenB) from comment #10)
> (FWIW, while the warning is important, brevity is key here,

I'm not so sure. We can assume that users will be familiar with the warning when it shows up as part of their normal workflow, so making it clear what's going on the first time they see the warning seems more important to me. Adding different icons or simple illustrations for the variations of the warning could help further...
(In reply to Meridel from comment #8)
> Hi Amy,
> 
> Can you help me understand the interactions that need copy? I see 4-8
> strings are mentioned. 
> 
> Can you please define and describe each distinct interaction? This will help
> me develop the content.
> 
> Thanks!

My understanding is that we need strings for the following:

1. Quit with session restore on
2. Quit with session restore off
3. Close multiple tabs with session restore on
4. Close multiple tabs with session restore off

Gijs can you confirm?
Flags: needinfo?(mjaritz)
Flags: needinfo?(gijskruitbosch+bugs)
Flags: needinfo?(amlee)
(In reply to Amy Lee [:amylee] UX from comment #12)
> (In reply to Meridel from comment #8)
> > Hi Amy,
> > 
> > Can you help me understand the interactions that need copy? I see 4-8
> > strings are mentioned. 
> > 
> > Can you please define and describe each distinct interaction? This will help
> > me develop the content.
> > 
> > Thanks!
> 
> My understanding is that we need strings for the following:
> 
> 1. Quit with session restore on
> 2. Quit with session restore off
> 3. Close multiple tabs with session restore on
> 4. Close multiple tabs with session restore off
> 
> Gijs can you confirm?

Well, right now we have 2 strings:

a) "You are about to close X tabs. Are you sure you want to continue."
And:
b) "You are about to close X windows with Y tabs. Are you sure you want to continue."

For the "closing multiple tabs" case (3 and 4 in your list), we never need the 'windows' case, as all tabs will be in the same window anyway. But I don't think we need 2 different strings there for the session restore pref either, as we will never automatically restore the closed tabs.

It is *possible* to *manually* restore the tabs using 'undo close tabs' and/or 'undo close window', and we may want to take the opportunity to e.g. offer a link to docs about that. However, the language doesn't need to differ based on whether the user has Firefox set to automatically restore the session on start or not because that won't affect the outcome (the user would need to manually restore the tabs/windows either way, because they're not quitting, only closing some tabs.

I think for (1) and (2) in your list, we'll need language for both the (a) and (b) case that I referenced. So I think that's 5 strings - 1a, 1b, 2a, 2b, and just one string for 3/4.

Does that make sense or am I missing something?

Separately, Amy, do you think we should have iconography or something else to more clearly distinguish the quit-with-sessionrestore-enabled (1) case from the multiple-tabs (3/4) case?
Flags: needinfo?(gijskruitbosch+bugs) → needinfo?(amlee)
Duplicate of this bug: 1507602
Apologies,but I am having some trouble following the number of *distinct* strings that are needed. Perhaps the SUMO article we reference should cover how to manually restore tabs so we do not need differentiated strings for all of the cases above? So, right now my understanding is that there are just two strings, with a recommendation to include instructions on manual tab restore in the SUMO article: https://docs.google.com/document/d/1O1w_5HMVW-taDZ0dWKfoeWj8aMHBkRx9lzLd6j9S8ag/edit
Flags: needinfo?(gijskruitbosch+bugs)
(In reply to Meridel from comment #15)
> Apologies,but I am having some trouble following the number of *distinct*
> strings that are needed. Perhaps the SUMO article we reference should cover
> how to manually restore tabs so we do not need differentiated strings for
> all of the cases above? So, right now my understanding is that there are
> just two strings, with a recommendation to include instructions on manual
> tab restore in the SUMO article:
> https://docs.google.com/document/d/1O1w_5HMVW-
> taDZ0dWKfoeWj8aMHBkRx9lzLd6j9S8ag/edit

I've responded in the doc. In terms of the different strings, the crux is that when the user (for instance) brings up the context menu on a tab, clicks "close tabs to the right" or "close other tabs", and we show a "you're closing N tabs" dialog, we shouldn't have any text about session restore, because Firefox isn't actually quitting (so whether automated session restore is enabled doesn't matter). Does that help?
Flags: needinfo?(gijskruitbosch+bugs) → needinfo?(mwalkington)
Thanks, that helps. I've updated the copy doc based on this information. Please take a look.

Since SUMO article already covers manual restore of tabs, I don't think we need additional copy in the dialogue itself.
Flags: needinfo?(mwalkington) → needinfo?(gijskruitbosch+bugs)
Duplicate of this bug: 1507931
I’m confused with what history is behind this bug and #550559 but I would like to voice my opinion on the current plan:

As a Windows user, I am not concerned with accidentally closing my browser window. When I close it, it’s very likely that I am doing that intentionally. Since I have session restore enabled, I conciously know that closing the browser will suspend my tabs and the browser will be closed. And when I reopen the browser, my tabs will be restored.

In this case, I do _not_ want to see a warning. With enabled session restore, I am perfectly aware of what is going on and what the browser will do. It is clear to me what the browser will look like when I reopen it later.

When I have multiple windows, and I close a window, the browser is _not_ exiting. So even with session restore (and the ability to reopen the window), I am losing some information if I close the window, so I _do_ want to see a warning message that asks me to confirm this.

The most common situation I actually got that warning in the past is because I was attempting to close my browser when there was still some popup or otherwise hidden window open that I forgot about. So the warning protects me from losing the state in my 200+ tab window.


This was how it has been in the past, and I am very used to this behavior, so are probably a lot of other users. So when the change first appeared on Nightly, showing a warning on _any_ close, a warning that before only appeared when there was actual loss, I was very concerned that session restore was simply broken (I am using Nightly after all, so that’s possible). In fact, I am *still* not trusting that message to properly restore my browser state, so I am not closing the browser on my machines at the moment (either I kill the process, or I shut down the computer with Firefox still open).

I would very much suggest that you restore the previous behavior, or at least allow me to configure it in some way. I do want to get notified when closing a window with tabs without closing the browser (removing the window from the session); I do not want to get notified when closing a window _while_ closing the browser (keeping the window in the session, subject for session restore). I consider the current behavior _very_ disruptive though.
(In reply to Patrick Westerhoff from comment #19)
> I do want to get notified when closing a window with tabs without closing the 
> browser (removing the window from the session); I do not want to get notified 
> when closing a window _while_ closing the browser

Go to about:config and set browser.warnOnQuit = false.

> a warning that before only appeared when there was actual loss

There is no loss here though, you can still go to the menu and Restore Previous Session. There is only loss if create some tabs and forget to restore the previous session before quitting so it gets overwritten.
Attached image close tabs spec.png (obsolete) —
Hi, 

I've attached a spec to more clearly outline the new strings in each scenario. I've bolded the copy for session restore enabled/disabled to highlight the differences. Let me know if you have any questions.
Flags: needinfo?(amlee)
This change has confused and annoyed me right from the point when it landed in Nightly. And till now I didn't know that it was intended and expected it to just be a bug that will be fixed.
I'm actually in the exact same situation as comment #19.
In the past week I have always quit the firefox process manually, because I didn't know what to do as it was the same message as when I close and loose all my tabs.


I now know that i can manually set "browser.warnOnQuit" to false and "browser.warnOnClose" to true to fix this.
But it's not really that nice that a user has to read bug reports for half an hour and follow links to other bug reports just to find a comment in there mentioning that you can do that.

Users who don't dig into about:config or who don't waste time to read bug reports about the issue won't know this.

They will be stuck with either one annoying warning too much. Or one very important warning too little.

And yes I know that I can restore manually via history->recently closed windows.
But I still prefer to neither have a annoying warning everytime nor to have the OH FOOF! moment when I accidentally close a window while not getting a warning.
Assignee: nobody → gijskruitbosch+bugs
Status: NEW → ASSIGNED
Flags: needinfo?(gijskruitbosch+bugs)
Duplicate of this bug: 1510364
Attachment #9026196 - Attachment is obsolete: true
I've attached an updated spec with final copy for the quit/closing dialogues. Let me know if you have any questions.
I think these are great.

Just one little request: Can you remove "Turn on session restore in preferences" ? The user might have deliberately chosen to turn it off, because it has severe privacy implications: Google and co can track me essentially forever (unless I explicitly delete cookies). So, we should not recommend this setting - it's simply a user choice. A user who has made this choice would then constantly see our recommendation to turn on session cookies. This is not the message we should give.
(In reply to Ben Bucksch (:BenB) from comment #26)
> I think these are great.
Me too.

> Just one little request: Can you remove "Turn on session restore in
> preferences" ? The user might have deliberately chosen to turn it off
In that case, the user can just ignore the link and not click it.
So I don't see any disadvantage in keeping it, and it can be helpful for users who wouldn't know where to configure this.

> because it has severe privacy implications: Google and co can track me
> essentially forever (unless I explicitly delete cookies).
Can you clarify this? How does session restore allow tracking? How are cookies related?

> So, we should not recommend this setting - it's simply a user choice.
If "Turn on Restore previous session" may make people believe it's somehow recommended, maybe a more neutral "Configure Restore previous session" or similar could be used instead? And it could make sense to also include it when session restore is on.
> Can you clarify this? How does session restore allow tracking? How are cookies related?

Bug 530594, bug 650298 etc.
"Session restore" leads to an "eternal session", you stay logged in even after you quit the browser, and it's not at all obvious to users.

> the user can just ignore the link and not click it.

Remember that this appears on every quit. If I have explicitly chosen to not have this on, for privacy reasons, and Firefox asks me every time I quit the browser to enable the preference, as if something was broken, I think that's way too pushy.

We should not asking for options that *decrease* privacy or security.
(In reply to Oriol Brufau [:Oriol] from comment #27)
> > So, we should not recommend this setting - it's simply a user choice.
> If "Turn on Restore previous session" may make people believe it's somehow
> recommended, maybe a more neutral "Configure Restore previous session" or
> similar could be used instead? And it could make sense to also include it
> when session restore is on.

Or "Note: session restore settings can be configured [here]".
Depends on: 1510557
I've moved the implementation of the link in the dialog to bug 1510557 because it complicates fixing this bug, and our window for 65 isn't very big anymore.
I've put up a patch. I think that basically covers the perceived regression from bug 550559. There are still some loose ends, though. Particularly:

0) as already noted, I haven't implemented a more specific warning (your tabs won't be restored) + link to the option in the 'quit' dialog for the non-sessionrestore case. Doing so is covered by bug 1510557. This isn't really related to bug 550559 directly as that was about the case where sessionstore is enabled.

1) The two checkboxes in the prefs can IMO be confusing if you enable session store, turn off warning for closing tabs (from the dialog when closing a window or otherwise closing more than 1 tab), and turn on warning for quitting underneath the session store pref. You end up with 2 checkboxes that both mention warning when quitting, one ticked and one unticked. I don't really know what the right solution to that problem is. Per discussion with UX, we *do* want to show the quit warning in that case (so the sessionstore-specific pref "overrides" the multiple-tabs pref) but I don't know how to make that clear in the UI, because the multiple tabs pref still controls whether we show a warning when closing a window / multiple tabs (but not quitting).

2) The language for the sessionrestore + quit case implies all windows will be restored. We don't restore private windows. Should we add a variant to that dialog that says "(except tabs from private windows)" (only when the user has private windows open) ? (Ni for this).
Note: I don't think this is critical and I think landing this fix before Monday's soft freeze (because l10n changes) *is*, so if I don't hear in time I might land in the current form and move the private browsing issue to a separate follow-up bug.
Flags: needinfo?(amlee)
(In reply to :Gijs (he/him) from comment #32)
> I've put up a patch. I think that basically covers the perceived regression
> from bug 550559. There are still some loose ends, though. Particularly:
> 
> 0) as already noted, I haven't implemented a more specific warning (your
> tabs won't be restored) + link to the option in the 'quit' dialog for the
> non-sessionrestore case. Doing so is covered by bug 1510557. This isn't
> really related to bug 550559 directly as that was about the case where
> sessionstore is enabled.
> 
> 1) The two checkboxes in the prefs can IMO be confusing if you enable
> session store, turn off warning for closing tabs (from the dialog when
> closing a window or otherwise closing more than 1 tab), and turn on warning
> for quitting underneath the session store pref. You end up with 2 checkboxes
> that both mention warning when quitting, one ticked and one unticked. I
> don't really know what the right solution to that problem is. Per discussion
> with UX, we *do* want to show the quit warning in that case (so the
> sessionstore-specific pref "overrides" the multiple-tabs pref) but I don't
> know how to make that clear in the UI, because the multiple tabs pref still
> controls whether we show a warning when closing a window / multiple tabs
> (but not quitting).

I'm wondering if we should remove "quitting" and only say "Warn me when closing multiple tabs" and keep the checkbox under session restore as "warn me when quitting the browser". NI Meridel for her input.  
 
> 
> 2) The language for the sessionrestore + quit case implies all windows will
> be restored. We don't restore private windows. Should we add a variant to
> that dialog that says "(except tabs from private windows)" (only when the
> user has private windows open) ? (Ni for this).
> Note: I don't think this is critical and I think landing this fix before
> Monday's soft freeze (because l10n changes) *is*, so if I don't hear in time
> I might land in the current form and move the private browsing issue to a
> separate follow-up bug.

An alternative is to mention that session restore is not enabled in private browsing when the user opens a private window instead of having more copy in preferences. NI Meridel for her input.
Flags: needinfo?(amlee) → needinfo?(mwalkington)
Pushed by gijskruitbosch@gmail.com:
https://hg.mozilla.org/integration/autoland/rev/b9dcfde98e47
use separate pref for warning on quit with sessionstore enabled, r=jaws,flod
https://hg.mozilla.org/mozilla-central/rev/b9dcfde98e47
Status: ASSIGNED → RESOLVED
Closed: 10 months ago
Resolution: --- → FIXED
Target Milestone: --- → Firefox 65
Thank you, everybody - Randell, Gijs, Amy, Meridel - for trying to find a good solution here!
Release Note Request (optional, but appreciated)
[Why is this notable]: Changes to long-standing quit behaviors
[Affects Firefox for Android]: no
[Suggested wording]: You can now configure Firefox to warn you when exiting, even when Firefox automatically restores your session.
[Links (documentation, blog post, etc)]: n/a

Note: this obsoletes the relnote for bug 550559.
relnote-firefox: --- → ?
Depends on: 1511819
Depends on: 1511820
I've split off the remaining issues into their own bug and moved the needinfos for Meridel there, so clearing this one. :-)
Flags: needinfo?(mwalkington)
Duplicate of this bug: 1510199
Flags: qe-verify+

Tested using Firefox 65.0b9 on Ubuntu 16.04 x64, macOS 10.13 and Windows 10 x64. The option to receive a notification even if the session restore preference is checked works as designed. The user is notified about the number of tabs and windows he is about to close with session restore and the warn preference on.

Please note that the content of the notification will be check in the separate issues submitted:

  • 1510557 (for the link for session restore).
  • 1511820 (about private browsing windows that won't be restored, implemented only in Fx66).
Status: RESOLVED → VERIFIED
Flags: qe-verify+
Duplicate of this bug: 1519611
You need to log in before you can comment on or make changes to this bug.