Can you repro this with the camera application?
On May 1, 6:22 am, blindfold seeingwithso...@gmail.com wrote:
I found that the old bug reported
inhttp://code.google.com/p/android/issues/detail?id=1578
where only a power cycle brings back the camera still persists with
the official Cupcake
Probably not as it may be load dependent, but once it happens with my
own app the built-in camera application also no longer works. It's
like the camera lock is no longer available to any app at that point,
including the one that last used it.
Regards
On May 1, 5:00 pm, Dave Sparks
Once the camera is in this state, Cupcake's built-in camera app gives
in LogCat
05-01 21:48:03.941: DEBUG/CameraService(35): Connect E from
ICameraClient 0x46290
05-01 21:48:03.951: DEBUG/CameraService(35): new client (0x46290)
attempting to connect - rejected
05-01 21:48:03.951:
Are you using the camera unlock() method? There is a known issue where
if the application calls unlock() and then exits for any reason, it
can leave the camera service in a state where it will reject any
subsequent attempts to reconnect.
On May 1, 12:57 pm, blindfold seeingwithso...@gmail.com
On May 1, 9:57 pm, blindfold seeingwithso...@gmail.com wrote:
OK, moving from Proguard 4.1 to 4.3 seems to bring much improved
stability as far my current testing goes, although I still managed to
get the dreaded camera lock once after a stressful transition
(quitting my app while it was
Is it advisable to add a Thread.sleep(1000L) after camera.release() to
make life easier on the camera service to prevent the unbound camera
lock situation? Of course that is not a fix, but I'll go with whatever
gives me good stability with Cupcake.
On May 1, 11:09 pm, blindfold
No I do not use the camera unlock() method anywhere. In fact I did not
even know this method existed. Seems undefined for the type Camera.
Did you perhaps mean camera.release()? I do use that, to be polite
against other camera apps when going in paused state, and when
quitting. Any
7 matches
Mail list logo
