shutdownTimeout
is reached because actionhero couldn't stop in time, we now throw an error rather than just simply process.exit(1)
. This will present the error to stdout properly and be caught as an uncaughtError
by actionhero startCluster
to be logged.api.cache.pop
, if there are no results, always return undefined
(some redis adaptors would return null)