This afternoon I went hunting for the source of UNL's failed pipelines. We have
deployed our capture agents configured to encode with x264 and use mp4
containers. Unfortunately, we have been experiencing problems with this setup.
I discovered a misconfiguration that caused the capture agent to default to
mpegps container, which is not technically capable of supporting h264. When I
corrected this mistake, I found that while the pipeline seemed to function, the
output was always corrupt.
It seems that since 1.0, we have been incorrectly shutting down our capture
pipeline. In most cases, the files are not messed up enough to cause any
significant hiccups, especially with mpeg2. But the mpeg4 files are completely
ruined.
I have created a bug indicating the issue, and consider it a release blocker
for us here at UNL. But our due to our lengthing 1.1 QA period, we may simply
accept that we have to stick to the happy path and use mpeg2 in the mean time.
http://opencast.jira.com/browse/MH-7593
Micah
_______________________________________________
Matterhorn mailing list
[email protected]
http://lists.opencastproject.org/mailman/listinfo/matterhorn
To unsubscribe please email
[email protected]
_______________________________________________
