399108 - Need an I'm-not-dead-yet message when minefield takes a while to close

Status: NEW

(Toolkit :: Startup and Profile System, defect)

Description

[Ed Hume]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9a9pre) Gecko/2007100804 Minefield/3.0a9pre
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9a9pre) Gecko/2007100804 Minefield/3.0a9pre

When I shut down minefield it takes about 25 seconds after the browser is closed until firefox.exe is no longer present in my task manager. This means that restarts take 30 seconds. It also means that if I shut down minefield I cannot restart it or start Firefox until I have waited that 25 seconds.

Firefox should leave an "I'm not dead yet" message until it is done doing whatever it does as it shuts down - a reverse splash screen, if you will.

Reproducible: Always

Steps to Reproduce:
1. Close Minefield.
2. Try to start it again or
3. Try to start Firefox.
Actual Results:  
There is an error message to the effect that another instance of Firefox is running.

Expected Results:  
Immediate restart.

If not that, some indication that Minefield is still active.

My places.sqlite file is 5318KB. 

Minefield does not seem to handle large bookmark collections very well.

Comment 1

[Jesse Ruderman]

It would be even better if it didn't take so long to shut down ;)

Comment 2

[Ria Klaassen (not reading all bugmail)]

Sometimes I even get the "Close Firefox..." message twice when I am in a hurry to start Firefox again after closing it. 

Comment 3

[Ed Hume]

I just received an email listing this bug as FIXED - WORKSFORME.

I then timed fx3. From when I clicked on the X to close the browser, it took 36 seconds for the firefox.exe listing to disappear from the Task Manager. 

It is true that the Firefox listing will disappear from the Applications tab of the Task Manager, but it is the firefox.exe listing in the Processes tab that tells us the app is still there.

This bug is not fixed.

Comment 4

[Jesse Ruderman]

You misread your bugmail.  This bug was not worked as WORKSFORME; bug 338884 was.

Comment 5

[Ed Hume]

Bug 407981 is a dupe of this bug. Bug 415201 covers the performance aspect relating to this bug. I would like to recommend that all further discussions occur in those bugs.

Comment 7

[Thomas Bertels]

(In reply to comment #5)

Bug 407981 seems focused on the Places issue, when this bug seems to be about the fact that we shouldn't have a message stating that Firefox is dead when it's not.

See http://mxr.mozilla.org/firefox/source/toolkit/xre/nsAppRunner.cpp#1648 for the code about it.

Comment 8

[James May [:fowl2]]

Wouldn't clearer notification on the less common case of the creation of a new instance be better than an always-on bastion of our less-than impressive shut-down performance?

Comment 9

[geko2u]

Attachment: Firefox 3.5.3 official exe

Comment 10

[Thomas Bertels]

geko2u> Could you please attach the code ?

Comment 11

[Thomas Bertels]

Comment on attachment 405236 [details]
Firefox 3.5.3 official exe

It's actually the official exe from Firefox 3.5.3.

By the way, the bug that introduced the problematic message is  bug 253950.

Comment 12

[dan_500]

from bug #447825 :
I can't start firefox shortly after quit. There should be a message "Wait while
firefox is restarting" and may a status bar [======....]. After successful quit
firefox should start instantly.