Bug 178357
| Summary: | iOS locks up when in portrait mode within WebRTC session | ||
|---|---|---|---|
| Product: | WebKit | Reporter: | Mike Block <mikeblock> |
| Component: | WebRTC | Assignee: | Nobody <webkit-unassigned> |
| Status: | RESOLVED DUPLICATE | ||
| Severity: | Critical | CC: | youennf |
| Priority: | P2 | ||
| Version: | Safari 11 | ||
| Hardware: | iPhone / iPad | ||
| OS: | iOS 11 | ||
Mike Block
While inside an RTC session, if I the device is in portrait mode, playback works for about 5 seconds and then the display becomes frozen and non-responsive.
The local stream is still delivered to the connected remote peers while the display is locked, at least for a minimum of 30 seconds after the device locks up.
There is nothing in the debug console reporting an issue. Not sure what events would be raised when this issue is triggered.
This is a consistent and repeatable event. I'm testing on an iPad Mini ME277C/A running iOS 11.0.3 (15A432)
- It seems from other reports that the lockup may release after several minutes on it's own.
- Tests have been conducted in a closed loop (i.e. all devices in the same physical network location).
- Once connected, there is no additional data being passed between the device and server outside the RTCPeerConnection objects themselves (i.e. no other TCP traffic)
If the screen is not rotated, no lockup seems to occur, but longer tests would be required to confirm this is the case.
Other developers seem to have suggested that cropping the display of a video inside an HTML container DIV using CSS is the reason for the issue.
| Attachments | ||
|---|---|---|
| Add attachment proposed patch, testcase, etc. |
Mike Block
Seems this is a duplicate of this ticket:
https://bugs.webkit.org/show_bug.cgi?id=176439
Hopefully having the entire OS lock up because of a CSS styling issue is considered worthwhile fixing. I'll close this ticket and follow the other one.
Mike Block
Sorry - leaving this as status new because resolved is not accurate.
youenn fablet
(In reply to Mike Block from comment #1)
> Seems this is a duplicate of this ticket:
>
> https://bugs.webkit.org/show_bug.cgi?id=176439
I would think so.
> Hopefully having the entire OS lock up because of a CSS styling issue is
> considered worthwhile fixing.
I would think so too!
Alexey Proskuryakov
Marking as duplicate per the above comments.
*** This bug has been marked as a duplicate of bug 176439 ***