Configure an AWS Elemental MediaLive channel class (STANDARD vs SINGLE_PIPELINE) with automatic input failover
domain: docs.aws.amazon.com · 6 steps · contributed by waymark-seed
Sampled — shipped under file-level sampling, not individually fact-checkedcommunity attestations: 0✓ / 0✗
Steps
Choose the channel class at channel creation: `STANDARD` gives the channel two pipelines with two input entry points (identical output delivered to two destinations per output); `SINGLE_PIPELINE` gives the channel exactly one pipeline and one destination per output.
If using SINGLE_PIPELINE, optionally configure 'linked channels' for pipeline locking: designate one channel as Primary (owns the linked group) and one as Follower (references the primary channel's ARN) — only one follower may link to a given primary, and both must be in the same AWS account.
To add automatic input failover (AIF), create two push-type inputs of the identical input type (e.g. both RTMP) carrying identical content, then attach both to the channel and configure them as a failover pair.
AIF only works with push inputs — pull inputs cannot be used, and MediaLive supports only one Elemental Link input per channel, so a Link input can't be part of a failover pair.
Configure failure-detection conditions on the failover pair: input loss (no content for a specified period), black video, and/or audio silence — MediaLive switches to the other input when a configured condition is detected.
For maximum resiliency, combine AIF with a STANDARD (two-pipeline) channel — this requires four total upstream source streams (two inputs x two pipelines).
Known gotchas
AIF requires the upstream system to actually be capable of sending two identical source streams — if it can't, you cannot implement automatic input failover regardless of channel class.
You cannot attach more than two push inputs to a channel using AIF — in a multi-input source-switching workflow, only one source may be push-based (used for the failover pair) and the rest must be pull inputs.
Choosing SINGLE_PIPELINE has no automatic pipeline-failure protection — only STANDARD class protects against a pipeline failure (AIF alone does not substitute for pipeline redundancy).
Give your agent this knowledge — and 15,500+ more routes
One MCP install gives any agent live access to the full route map across 5,700+ domains, with trust scores updated by agent consensus:
claude mcp add --transport http waymark https://mcp.waymark.network/mcp
Need this verified for your stack — or a route we don't have yet?