Troubleshooting
Pipeline locking ensures that the pipelines are frame accurate with each other, in the output groups where MediaLive performs pipeline locking.
If you or the operator at the downstream system notices that the pipelines are not synchronized, you can perform the following troubleshooting.
General troubleshooting
These troubleshooting steps apply to all pipeline locking methods:
-
Make sure that MediaLive supports pipeline locking for the type of input in your channel.
-
Make sure that the affected outputs are eligible for pipeline locking. Pipeline locking applies only to specific types of outputs.
-
Make sure that you changed the Framerate control so that it is not Initialize_from_source.
-
Check the ComplexFRCPresent CloudWatch metric. A value of 1 indicates that Medialive is performing complex framerate conversion and is not attempting to lock pipelines. Pipeline locking only supports simple framerate conversions.
-
If the framerate within the source changes, it's possible that MediaLive can't perform pipeline locking for the duration because for that section of video, there is no simple framerate conversion.
Troubleshooting timecode-based locking
If you are using epoch locking, or pipeline locking with the source timecode method (the default), check the following in addition to the general troubleshooting steps:
-
Make sure that the input source has an embedded timecode.
-
If you chose epoch-locking mode, make sure that the embedded timecode is within 2 minutes of epoch time.
-
If an input source has sections where there is no embedded timecode, MediaLive stops performing frame-accurate pipeline locking. MediaLive automatically falls back to performing approximate pipeline locking. Whenever the embedded timecode reappears, MediaLive resumes frame-accurate pipeline locking.
-
Make sure that you remembered to set up segmentation markers in a UDP output group. For the other supported output groups, you don't need to worry about this because their outputs are always segmented.
-
Make sure that you set up the segmentation marker type that your downstream system expects.
Troubleshooting video aligned locking
If you are using video aligned locking (Pipeline locking method set to VIDEO_ALIGNMENT) and experience synchronization issues, check the following in addition to the general troubleshooting steps:
-
Verify that the current input type is compatible. HLS, RTMP_PULL, and file inputs cause video aligned locking to run in open loop mode (unlocked). See Requirements for video aligned locking.
-
Check the InputVideoAligned CloudWatch metric. A value of 1 indicates pipeline locking has successfully aligned the input video content between pipelines. If the value is 0:
-
Ensure both pipelines are receiving the same video content. Video aligned locking compares visual signatures between encoders and cannot lock if the content differs.
-
-
Check the PipelinesLocked CloudWatch metric. Video aligned locking reports its locked status through this same metric. A value of 1 indicates successful synchronization.
-
If synchronization is intermittent, verify that your network connectivity to both pipelines is stable. Visual signature comparison requires consistent video delivery to both encoders.
