As a workaround for this issue, Sitka has switched from Apache Websockets to websocketd, using Bill's branch from bug 1777180. It's working well for us: no spinning websockets, segfaults, or other problems that I know of.
As a workaround for this issue, Sitka has switched from Apache Websockets to websocketd, using Bill's branch from bug 1777180. It's working well for us: no spinning websockets, segfaults, or other problems that I know of.