@markusthoemmes one thing we can do here is rely on `host-connection-pool.idle-timeout`, instead of explicitly calling `pool.shutdown` (still not sure how to get the materialized instance of the existing pool...). If we set that `idle-timeout` to the `pause-grace` value - the pool should go away around the time that pause occurs. If it happens after the pause (since we loose control to wait for it during pause processing), hopefully that is ok? If it is not ok, we could always set the timeout to `pause-grace - 10ms` (and in some cases the pool will be recreated before pause ever occurs). This is definitely less precise, but it may be an option if the explicit pool shutdown is not accessible.
[ Full content available at: https://github.com/apache/incubator-openwhisk/pull/3976 ] This message was relayed via gitbox.apache.org for [email protected]
