Ettore Di Giacinto d498480c65 docs: design model failover chains ...

A model served from a remote upstream has no local fallback when that
upstream is unhealthy, so clients reimplement failover and lose realtime
context when they switch endpoints.

Define a failover block on model configs: an ordered list of targets with
active probes, in-request retry before the response is committed,
fail-back with hysteresis, and switch events over REST, SSE and the
realtime socket.

Assisted-by: Claude:claude-opus-5-5

2026-09-26 14:51:26 +00:00

..

2026-08-21-configurable-copy-buffer-design.md

feat(xio): make copy buffer size configurable (#11660)

2026-08-21 16:14:26 +02:00

2026-08-21-distributed-model-config-revisions-design.md

fix(distributed): keep model replicas config-consistent (#11664)

2026-08-22 22:44:03 +02:00

2026-08-21-distributed-staging-operations-design.md

fix(distributed): keep staging operations stable (#11663)

2026-08-21 20:08:47 +02:00

2026-08-21-scheduling-rule-editing-node-labels-design.md

feat(ui): edit scheduling rules in place (#11667)

2026-08-22 00:51:48 +02:00

2026-09-07-ephemeral-staging-retention-design.md

fix(distributed): bound ephemeral staging (#11924)

2026-09-08 18:13:40 +02:00

2026-09-07-exl3-gallery-design.md

feat(gallery): add EXL3 vllm.cpp models (#11912)

2026-09-07 20:55:42 +02:00

2026-09-26-model-failover-chains-design.md

docs: design model failover chains

2026-09-26 14:51:26 +00:00