On 11/28/2018 02:33 AM, Daniel Ruggeri wrote: > Hi, Jim; > I'm coming up empty on a search against the Google-machine for SURVEY > protocol. Any pointers you can share? > > I'm also curious what your thoughts are about dealing with growth of the > worker pool beyond the control of the proxy admin. For example, if I configure > a balancer for 10 nodes, I have an idea of tuning parameters I might set > differently than if the number of backends is in the tens of servers. > -- > Daniel Ruggeri
Mod_cluster solves this with configuration messages from worker nodes. Httpd with mod_cluster configures balancers (as in mod_proxy balancer lingo) according to CONFIG messages from workers. So if you are spinning up a new batch of workers, you might tell them to configure balancer differently, different timeout or retry values for example. Cheers K. > > On November 27, 2018 5:23:25 AM CST, Jim Jagielski <[email protected]> wrote: > > In the coming week or so, I will be committing my load balance, > load determination and discovery work to a sandbox trunk. Many > people have asked for more info, so here we go. > > Basically, this new feature uses nanomsg (nng) to implement the > SURVEY protocol between workers (nodes) and the front end server. > The proxy server will send out MEMBERSHIP and STATUS surveys, that > nodes can respond to; thus new nodes can automagically add themselves > as they come up as well as remove themselves when they restart or > shutdown via MEMBERSHIP. STATUS can be used to provide real-time > load on each node, which will be sent via a __packed__ struct, > ala NTP and other common usage patterns, in which 32bit fields are > converted from endian-ness to network and back. That STATUS info can > be used for "intelligent" load balancing and will provide some > typical server metrics such as memory, # of cpus, etc... > Michal Karm Babacek -- Sent from my Hosaka Ono-Sendai Cyberspace 7
