After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 748893 - RTSP clock jumps after OSX sleep
RTSP clock jumps after OSX sleep
Status: RESOLVED DUPLICATE of bug 758012
Product: GStreamer
Classification: Platform
Component: dont know
git master
Other Mac OS
: Normal normal
: git master
Assigned To: GStreamer Maintainers
GStreamer Maintainers
Depends on:
Blocks:
 
 
Reported: 2015-05-04 14:07 UTC by Vivia Nikolaidou
Modified: 2015-12-07 12:22 UTC
See Also:
GNOME target: ---
GNOME version: ---


Description Vivia Nikolaidou 2015-05-04 14:07:47 UTC 
Only tried on OSX so far.

Sender pipeline: ./test-launch "videotestsrc is-live=true pattern-18 ! timeoverlay font-desc=\",48\" ! video/x-raw,width=1280,height=720,framerate=30/1 ! x264enc ! h264parse ! rtph264pay name=pay0 pt=96"
Receiver pipeline: Just play the above using gst-play.

Steps to reproduce: Start the above, go to lunch or otherwise leave the Macbook intact for maybe half an hour, come back, wake up the Macbook.

Symptom: Both the bouncy ball from videotestsrc and the clock from timeoverlay will be jumping forward - they play back for maybe about a second, then jump forward for a couple of seconds, etc etc.

Lame screen capture (from phone) available at http://ahiru.eu/~vivia/lunch_bug.mp4

Will test on Linux and report back.
Comment 1 Vivia Nikolaidou 2015-05-04 14:33:37 UTC 
!!! TYPO ALERT !!!

In the sender pipeline, instead pattern-18 it should be pattern=18 of course. So, for the sake of pasting, the correct pipeline is:

./test-launch "videotestsrc is-live=true pattern=18 ! timeoverlay font-desc=\",48\" ! video/x-raw,width=1280,height=720,framerate=30/1 ! x264enc ! h264parse ! rtph264pay name=pay0 pt=96"

It is the sleep mode on OSX that triggers the bug. Enabling the "Prevent computer from sleeping automatically when the display is off" option prevents the bug. Also it works fine on Linux.
Comment 2 Heinrich Fink 2015-12-04 15:19:26 UTC 
This might behave differently now since GstSystemClock uses mach_absolute_time (see #758012).
Comment 3 Sebastian Dröge (slomo) 2015-12-07 11:09:14 UTC 
Can someone test and confirm?
Comment 4 Heinrich Fink 2015-12-07 12:12:27 UTC 
Just did a quick test: works for me.
Comment 5 Sebastian Dröge (slomo) 2015-12-07 12:22:03 UTC 

*** This bug has been marked as a duplicate of bug 758012 ***