1036685 - [B2G][2.0][Browser] Out of memory issue when the user watch videos on NBA, NHL, NFL.com

Status: RESOLVED DUPLICATE of bug 862743

(Web Compatibility :: Site Reports, defect)

Description

[pete siphantong [:Petes]]

Attachment: nb.txt

Description:
When the user watch a video on nfl.com, the browser will crash 

Setup Steps:
1) Flame device is set to 273mb

Repro Steps:
1) Update a Flame device to BuildID: 20140708000322
2) Go to Browser
3) Go to either nba, nfl, or nhl.com
4) Tap video


Actual:
the browser crash

Expected:
the browser goes to the download/sign in page

Environmental Variables:
Device: Flame 2.0 
Build ID: 20140708000322
Gaia: e935f4ff190b76c70d9b2af8856c542a6e4a7546
Gecko: 3f9d7a3a0b7b
Version: 32.0a2 (2.0) 
Firmware Version: v122
User Agent: Mozilla/5.0 (Mobile; rv:32.0) Gecko/32.0 Firefox/32.0

Repro frequency: 100%
See attached: screenshot, logcat

Comment 1

[pete siphantong [:Petes]]

Attachment: 2014-07-09-16-02-24.png

Comment 2

[pete siphantong [:Petes]]

Please ignore the expected write up, correct expected: The browser plays the video 


This issue also reproduces on the Flame 2.1 (273mb), Flame 1.4(273mb),Flame Base Image v.122(273mb). The browser crash

Flame 2.1 (273mb)

Environmental Variables:
Device: Flame Master(273MB)
BuildID: 20140709040203
Gaia: 0f9f11d0a6dadb3ea27160204bbe911c1ad69a6f
Gecko: 196d05832e12
Version: 33.0a1 (Master) 
Firmware Version: v122
User Agent: Mozilla/5.0 (Mobile; rv:33.0) Gecko/33.0 Firefox/33.0

Flame 1.4(273MB)

Environmental Variables:
Device: Flame 1.4
Build ID: 20140709003002
Gaia: b0e9b4bdb39c5eb93a6783a34624ffc84f62b126
Gecko: acf704e54e19
Version: 30.0 (1.4)
Firmware Version: v122
User Agent: Mozilla/5.0 (Mobile; rv:30.0) Gecko/30.0 Firefox/30.0

Flame Base Image v.122(273mb)

Environmental Variables:
Device: Flame 1.3
Build ID: 20140616171114
Gaia: e1b7152715072d27e0880cdc6b637f82fa42bf4e
Gecko: e181a36ebafaa24e5390db9f597313406edfc794
Version: 28.0 (1.3)
Firmware Version: v122
User Agent: Mozilla/5.0 (Mobile; rv:28.0) Gecko/28.0 Firefox/28.0



------------------------------------------------------------------------------------------------------------------

This issue does not reproduce on the Buri 2.1, Open C 2.1, Flame 2.0(512mb), Buri 2.0, Open C 2.0, Buri 1.4, Open C 1.4. The browser plays the video 



Buri 2.1

Environmental Variables:
Device: Buri Master
Build ID: 20140709073020
Gaia: c394b7b4205b6f1a6ca44915fc08650f3ad127ec
Gecko: 2d88803a0b9c
Version: 33.0a1 (Master)
Firmware Version: v1.2device.cfg
User Agent: Mozilla/5.0 (Mobile; rv:33.0) Gecko/33.0 Firefox/33.0

Open C 2.1

Environmental Variables:
Device: Open_C Master
Build ID: 20140708040218
Gaia: 740faa5d0060fb218b407cf224330654ddf833a5
Gecko: 465280604ea6
Version: 33.0a1 (Master)
Firmware Version: P821A10V1.0.0B06_LOG_DL
User Agent: Mozilla/5.0 (Mobile; rv:33.0) Gecko/33.0 Firefox/33.0

Flame 2.0(512mb)

Environmental Variables
Device: Flame v2.0 
Build ID: 20140708000322
Gecko: https://hg.mozilla.org/releases/mozilla-aurora/rev/3f9d7a3a0b7b
Gaia: e935f4ff190b76c70d9b2af8856c542a6e4a7546
Platform Version: 32.0a2
Firmware Version: v122
User Agent: Mozilla/5.0 (Mobile; rv:32.0) Gecko/32.0 Firefox/32.0

Buri 2.0

Environmental Variables:
Device: Buri 2.0
Build ID: 20140709063007
Gaia: 1774027323bb072b4ebdfea9883572bcf2535c87
Gecko: 11b6493a7d8f
Version: 32.0a2 (2.0)
Firmware Version: v1.2device.cfg
User Agent: Mozilla/5.0 (Mobile; rv:32.0) Gecko/32.0 Firefox/32.0

Open C 2.0

Environmental Variables:
Device: Open_C 2.0
Build ID: 20140708000322
Gaia: e935f4ff190b76c70d9b2af8856c542a6e4a7546
Gecko: 3f9d7a3a0b7b
Version: 32.0a2 (2.0)
Firmware Version: P821A10V1.0.0B06_LOG_DL
User Agent: Mozilla/5.0 (Mobile; rv:32.0) Gecko/32.0 Firefox/32.0


Buri 1.4

Environmental Variables:
Device: Buri 1.4
Build ID: 20140709003002
Gaia: b0e9b4bdb39c5eb93a6783a34624ffc84f62b126
Gecko: acf704e54e19
Version: 30.0 (1.4)
Firmware Version: v1.2device.cfg
User Agent: Mozilla/5.0 (Mobile; rv:30.0) Gecko/30.0 Firefox/30.0

Open C 1.4

Environmental Variables:
Device: Open_C 1.4
Build ID: 20140709000201
Gaia: b0e9b4bdb39c5eb93a6783a34624ffc84f62b126
Gecko: acf704e54e19
Version: 30.0 (1.4)
Firmware Version: P821A10V1.0.0B06_LOG_DL
User Agent: Mozilla/5.0 (Mobile; rv:30.0) Gecko/30.0 Firefox/30.0

Comment 3

[KTucker [:KTucker][Inactive 3/4/2016]]

Could you please attach a video for this? You changed 1.3 tracking flag to affected yet I do not see this indicated in your regression check comment.

I don't think your product is correct either.

Comment 4

[Karl Dubost💡 :karlcow]

It doesn't seem to be a Web Compatibility issue.

Comment 5

[pete siphantong [:Petes]]

This issue does not occur on the Buri 1.3, the video plays on nfl.com, the video does not play in nhl.com because the format is incompatible and for nba.com the browser crashes going to the main site

Environmental Variables:
Device: Buri 1.3
Build ID: 20140709024000
Gaia: 23f55be856cef53c6604a6fe4aeb09061afbc897
Gecko: 601d27e413a9
Version: 28.0 (1.3)
Firmware Version: v1.2device.cfg
User Agent: Mozilla/5.0 (Mobile; rv:28.0) Gecko/28.0 Firefox/28.0

Comment 6

[KTucker [:KTucker][Inactive 3/4/2016]]

This issue appears to be a result of the 273mb of memory. This is a regression in regards to the user not being able to play videos from NFL.com. I don't think this issue should block release but i could see this issue impacting a lot of end users.

Comment 7

[Jason Smith [:jsmith]]

(In reply to ktucker from comment #6)
> This issue appears to be a result of the 273mb of memory. This is a
> regression in regards to the user not being able to play videos from
> NFL.com. I don't think this issue should block release but i could see this
> issue impacting a lot of end users.

That doesn't make sense to me. The fact that an issue is a regression & involves a basic use case with being unable to do something in web content is a critical issue. Please re-evaluate.

Comment 8

[KTucker [:KTucker][Inactive 3/4/2016]]

After reevaluating the impact of this issue,  nominating this 2.0? This is a regression from 1.3 and could affect many users.

Comment 9

[Jason Smith [:jsmith]]

(In reply to ktucker from comment #8)
> After reevaluating the impact of this issue,  nominating this 2.0? This is a
> regression from 1.3 and could affect many users.

There should also be a window requested here as well.

Comment 10

[Sotaro Ikeda [:sotaro]]

It seems a graphic related problem. The number of memory used by media framework seems basically same. graphic side have some regressions related to this recently.

Comment 11

[Jason Smith [:jsmith]]

Kevin is out, so redirecting to Peter.

Comment 12

[Peter Bylenga [:PBylenga]]

adding RegressionWindow-Wanted given the criteria reflected in Comment 8.

Comment 13

[Jason Smith [:jsmith]]

The flags are wrong here. If you see above, the tester indicated that this was reproducing on all applicable Flame configurations (including 1.3). That means this is not a regression.

(In reply to Karl Dubost :karlcow from comment #4)
> It doesn't seem to be a Web Compatibility issue.

I disagree. A site that is poorly designed for low memory devices is classified as a site issue, not a Firefox OS issue.

Comment 14

[Hallvord R. M. Steen [:hallvors]]

Comment 10 says "graphic side have some regressions related to this recently". Yes, if a page does something that's clearly non-optimal for the computing capacity of a mobile phone we could treat that as a tech evangelism issue - but if we're going to contact them we'd like to know 100% that it's not our fault (though some inefficiency or memory leak) and what fix we'd recommend.

I don't think this bug has either a thorough analysis or a suggested fix. Yet. So I agree with Karl that at this stage, it's not a Tech Evang bug.

Comment 15

[Jason Smith [:jsmith]]

(In reply to Hallvord R. M. Steen from comment #14)
> Comment 10 says "graphic side have some regressions related to this
> recently". Yes, if a page does something that's clearly non-optimal for the
> computing capacity of a mobile phone we could treat that as a tech
> evangelism issue - but if we're going to contact them we'd like to know 100%
> that it's not our fault (though some inefficiency or memory leak) and what
> fix we'd recommend.

This isn't a regression. If it isn't a regression, then we can't classify it as our fault.

> 
> I don't think this bug has either a thorough analysis or a suggested fix.
> Yet. So I agree with Karl that at this stage, it's not a Tech Evang bug.

I don't think you've presented any new data to prove that this is a gecko bug. This isn't a regression, which concludes that this is TE bug, not a gecko bug. Back to TE.

Comment 16

[Jason Smith [:jsmith]]

(In reply to Jason Smith [:jsmith] from comment #15)
> (In reply to Hallvord R. M. Steen from comment #14)
> > Comment 10 says "graphic side have some regressions related to this
> > recently". Yes, if a page does something that's clearly non-optimal for the
> > computing capacity of a mobile phone we could treat that as a tech
> > evangelism issue - but if we're going to contact them we'd like to know 100%
> > that it's not our fault (though some inefficiency or memory leak) and what
> > fix we'd recommend.
> 
> This isn't a regression. If it isn't a regression, then we can't classify it
> as our fault.
> 
> > 
> > I don't think this bug has either a thorough analysis or a suggested fix.
> > Yet. So I agree with Karl that at this stage, it's not a Tech Evang bug.
> 
> I don't think you've presented any new data to prove that this is a gecko
> bug. This isn't a regression, which concludes that this is TE bug, not a
> gecko bug. Back to TE.

comment 10 also I don't think applies here if this isn't a regression.

Comment 17

[Karl Dubost💡 :karlcow]

(In reply to Jason Smith [:jsmith] from comment #13)
> I disagree. A site that is poorly designed for low memory devices is
> classified as a site issue, not a Firefox OS issue.

This is not a Web Compatibility issue by your own working.
It might be an optimization issue. A Web compatibility issue is when a Web site has been created in a way that clearly exclude one or a few browsers. For example, a site working on Chrome Android and Safari iOS but not on Firefox Android, because CSS or JS have been done wrong and we *know* it could work on Firefox for Android.

There are plenty of Web sites out there which are NOT optimized in terms of performance for certain devices, we do not necessary contact them. Some sites for their own business reasons will choose Flash as a technology instead of HTML5, etc.

So to come back to the real question, is this "Out Of Memory" is created by a specific usage of JS which impacts Firefox specifically and is working only in other browsers. And then we can try to see if it's really:
1. a Tech Evangelism issue, aka "if your JS does this, it works the same everywhere".
2. OR a performance issue of Firefox in General (doesn't even have to be a regression) aka "Firefox really sucks at doing this kind of tasks".

It's what hallvord is saying by proper analysis.

Comment 18

[Karl Dubost💡 :karlcow]

s/by your own working/by your own wording/ in the previous comment.

Comment 19

[Karl Dubost💡 :karlcow]

# NBA
http://www.nba.com/ 
Redirection done. Working.
-> http://mi.nba.com/   
No issue

# NFL
http://www.nfl.com/
Receiving Desktop content.
https://bugzilla.mozilla.org/show_bug.cgi?id=862743 was an empty meta bug.
It never had a dependency we can mutate it as the bug for this issue
Just done.

# NHL
http://www.nhl.com/  
Receiving Desktop Content
The bug already exists.
    https://bugzilla.mozilla.org/show_bug.cgi?id=973354

Comment 20

[Jason Smith [:jsmith]]

(In reply to Karl Dubost :karlcow from comment #17)
> (In reply to Jason Smith [:jsmith] from comment #13)
> > I disagree. A site that is poorly designed for low memory devices is
> > classified as a site issue, not a Firefox OS issue.
> 
> This is not a Web Compatibility issue by your own working.

This is speculative - we should be proving this with hard evidence. For example, if you checked NHL.com, you would have noticed that a mobile site was rendering on IPhone, but not Firefox OS. That's a cut and dry compatibility issue.

> It might be an optimization issue. A Web compatibility issue is when a Web
> site has been created in a way that clearly exclude one or a few browsers.
> For example, a site working on Chrome Android and Safari iOS but not on
> Firefox Android, because CSS or JS have been done wrong and we *know* it
> could work on Firefox for Android.

*might* isn't the right way to approach this. We need to be *sure* something is a TE issue or not. NHL we already know is a compatibility issue for example from the statement above, so that issue is definitely a TE problem.

> 
> There are plenty of Web sites out there which are NOT optimized in terms of
> performance for certain devices, we do not necessary contact them. Some
> sites for their own business reasons will choose Flash as a technology
> instead of HTML5, etc.
> 
> So to come back to the real question, is this "Out Of Memory" is created by
> a specific usage of JS which impacts Firefox specifically and is working
> only in other browsers. And then we can try to see if it's really:
> 1. a Tech Evangelism issue, aka "if your JS does this, it works the same
> everywhere".
> 2. OR a performance issue of Firefox in General (doesn't even have to be a
> regression) aka "Firefox really sucks at doing this kind of tasks".
> 
> It's what hallvord is saying by proper analysis.

Then you should ask for what data you are missing in order to fill the gaps to fully determine if it's TE or not. A global triager's job is figure out the most likely suspected area a bug belongs too, but that means sometimes it might not have all of the data in the bug up front. The experts in a particular component have a responsibility for asking the questions to fill the gaps to fully confirm if the issue fits in the component or not. Throwing over the bug though randomly certainly isn't right way to get the problem resolved, as that's the equivalent to throwing out someone's hard work to write a bug out the door.

Comment 21

[Jason Smith [:jsmith]]

Actually, this is a dupe of bug 862743, not invalid. The bug as filed here was focusing on nfl.com.