666236 - Component returned failure code: 0x8000fff (NS_ERROR_UNEXPECTED) [nsIWebNavigation.goBack | nsIWebNavigation.goForward] @ browser.js

Status: VERIFIED FIXED

(Firefox for Android Graveyard :: General, defect)

Description

[Cristian Nicolae (:xti)]

Attachment: screenshot

Build id : Mozilla/5.0 (Android;Linux armv7l;rv:7.0a1)Gecko/20110621
Firefox/7.0a1 Fennec/7.0a1
Device: Motorola Droid 2
OS: Android 2.2

The following error occurs when you may want to go back to the previous website and the device back button/back arrow from the right panel doesn't perform the expected action at the first tap:

Error: uncaught exception: [Exception... "Component returned failure code: 0x8000fff (NS_ERROR_UNEXPECTED) [nsIWebNavigation.goBack]" nsresult: "0x8000fff (NS_ERROR_UNEXPECTED)" location: "JS frame :: chrome://browser/content/bindings/browser.js :: <TOP_LEVEL> :: line 146" data: no]

Comment 1

[Mark Finkle (:mfinkle) (use needinfo?)]

This happens all the time? If not, give more details about when you see this happen.

Comment 2

[Aaron Train [:aaronmt]]

I just hit the equivalent with goForward() on 06/22 Nightly

uncaught exception: [Exception... "Component returned failure code: 0x8000ffff (NS_ERROR_UNEXPECTED) [nsIWebNavigation.goForward]"  nsresult: "0x8000ffff (NS_ERROR_UNEXPECTED)"  location: "JS frame :: chrome://browser/content/bindings/browser.js :: <TOP_LEVEL> :: line 150"  data: no]

STR:
1. www.hardocp.com
2. Clicked on any of the top motherboard reviews
3. Tapped Back and Forward a couple times

Comment 3

[Cristian Nicolae (:xti)]

This issue doesn't occur on:
Build id : Mozilla/5.0 (Android;Linux armv7l;rv:6.0a1)Gecko/20110513
Firefox/6.0a1 Fennec/6.0a1

but it occurs on:
Build id : Mozilla/5.0 (Android;Linux armv7l;rv:6.0a1)Gecko/20110514
Firefox/6.0a1 Fennec/6.0a1
http://hg.mozilla.org/mozilla-central/rev/932a10564fbc

Possible range:
http://hg.mozilla.org/mozilla-central/pushloghtml?startdate=2011-05-13&enddate=2011-05-14+03%3A00

Comment 4

[Aaron Train [:aaronmt]]

Hi Cristian, can you get a range using build ID's?

Comment 5

[Aaron Train [:aaronmt]]

(In reply to comment #4)
> Hi Cristian, can you get a range using build ID's?

I mean changesets

Comment 6

[Cristian Nicolae (:xti)]

(In reply to comment #5)
> (In reply to comment #4)
> > Hi Cristian, can you get a range using build ID's?
> 
> I mean changesets

http://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=7fd948de62a6&tochange=932a10564fbc

Comment 7

[Doug Turner (:dougt)]

kevin, can you help here?  we need to know if (a) this still happens, and if so (b) can you reduce this range to once changeset?

Comment 8

[Aaron Train [:aaronmt]]

(In reply to comment #7)
> kevin, can you help here?  we need to know if (a) this still happens, and if
> so (b) can you reduce this range to once changeset?

This still happens 08/04 - to repro, just visit two sites, hit back and forward rapidly - check error console

Comment 9

[Mark Finkle (:mfinkle) (use needinfo?)]

Testing on the 08/04 Aurora I see the following:
* Back and Forward work without any errors if you allow the page to at least _start_ loading before tapping Back/Forward again
* Back and Forward produce errors in the console if you tap them in a rapid sequence. The page still tries to go back or forward though. It seems to actually work most of the time.

Therefore, I suggest we do nothing for this bug in Aurora unless someone can show that the app crashes or is somehow affected negatively. If the errors occur in Nightly as well (comment 2 makes it seem like it does), we can use a try/cacth to hide the errors from the error console.

In Fennec, back/forward are async operations. They do not block and happen instantly as on desktop (for now). We can't expect them to respond to rapid sequences, but we should not crash or corrupt the app either.

Comment 10

[Lucas Rocha (:lucasr)]

Attachment: (1/2) Don't pass argument to methods that don't use it

Comment 11

[Lucas Rocha (:lucasr)]

Attachment: (2/2) Only allow browser to go back/forward when those are valid operations

Comment 12

[Mark Finkle (:mfinkle) (use needinfo?)]

Comment on attachment 554883 [details] [diff] [review]
(2/2) Only allow browser to go back/forward when those are valid operations

Wait a minute. These are the same patches.

Comment 13

[Lucas Rocha (:lucasr)]

Attachment: (1/2) Don't pass argument to methods that don't use it

Oops, sorry about that :-P

Comment 14

[Naoki Hirata :nhirata (please use needinfo instead of cc)]

I ran into this on the nightly after going to https://www.adpweb.com.br/rhweb18/ then doing a google search on .gov.uk and hitting the back button

Mozilla/5.0 (Android; Linux armv7I; rv9.0a1) Gecko/20110822 Firefox/9.0a1 Fennec/9.0a1
Device: HTC Flyer 
OS: Android 2.3

Comment 15

[Ed Morley [:emorley]]

In my queue, which is being sent to try and then presuming green, I'll land on inbound later today.

Comment 16

[Naoki Hirata :nhirata (please use needinfo instead of cc)]

Can this be related to bug 680344 as well?

Comment 17

[Lucas Rocha (:lucasr)]

Naoki, it might be related. It's worth trying to reproduce bug 680344 with a build with this patch applied.

Comment 18

[Ed Morley [:emorley]]

http://tbpl.allizom.org/?tree=Try&usebuildbot=1&rev=e3f0e5b06e77

http://hg.mozilla.org/integration/mozilla-inbound/rev/5347b86eba43
http://hg.mozilla.org/integration/mozilla-inbound/rev/d9859ad8932b

Comment 19

[Marco Bonardo [:mak]]

http://hg.mozilla.org/mozilla-central/rev/d9859ad8932b
http://hg.mozilla.org/mozilla-central/rev/5347b86eba43

Comment 20

[Naoki Hirata :nhirata (please use needinfo instead of cc)]

Mozilla/5.0 (Android; Linux armv7I; rv9.0a1) Gecko/20110826 Firefox/9.0a1 Fennec/9.0a1
Device: HTC Flyer 
OS: Android 2.3