The radicle-ci-broker supports a single max_run_time config option. This is a global setting that applies to all adapter invocations by the broker.

As a CI node administrator, I want to service workloads that have very different "max runtimes": one repo might typically run in minutes, another may take hours. Clearly, setting max_run_time based on the latter would hurt cases where the former adapter is stuck - essentially blocking all CI builds on that repo for several hours.

It sounds like this is my only option right now, so that's what I'll go for, but I wanted to highlight this as it seems to me there's a per-repo setting missing here.

Link to zulip thread: https://radicle.zulipchat.com/#narrow/channel/452370-radicle-ci/topic/per-repo.20timeout/near/594756115