sccache: when scheduler is unreachable, sccache panics with "thread 'tokio-runtime-worker' panicked at 'Cannot start a runtime from within a runtime. This happens because a function (like `block_on`) attempted to block the current thread while the thread is being used to drive asynchronous tasks.', .cargo/registry/src/github.com-1ecc6299db9ec823/tokio-1.16.1/src/runtime/enter.rs:39:9

I’m using a local release build of sccache tip-of-tree (ec1457e8cb508c12d9f177074e20ca89f790489e). I’ve got a scheduler configured in my ~/.config/sccache/config, like so:

[dist]
scheduler_url = "http://10.0.32.3:10600"

My scheduler is usually offline, which is supposed to cause sccache to fall back to a local-only configuration (and indeed that’s what happens with e.g. the version of sccache that’s bundled with mozilla-central). But in my own local build of sccache trunk, when I build Firefox while my remote sccache scheduler is offline, I quickly get build failures due to sccache panicking, with logging like this:

[2022-04-19T20:00:45Z INFO sccache::server] Enabling distributed sccache to http://10.0.32.3:10600/

thread ‘tokio-runtime-worker’ panicked at ‘Cannot start a runtime from within a runtime. This happens because a function (like block_on) attempted to block the current thread while the thread is being used to drive asynchronous tasks.’, .cargo/registry/src/github.com-1ecc6299db9ec823/tokio-1.16.1/src/runtime/enter.rs:39:9

I’ll post a full backtrace in a bit.