EndpointInfo uses composition to provide info about the different transports, you can get the TCPEndpoint info by checking the underlying endpoint info, something like
getTCPEndpointInfo(const Ice::EndpointInfoPtr& info)
for(Ice::EndpointInfoPtr p = info; p; p = p->underlying)
Ice::TCPEndpointInfoPtr tcpInfo = Ice::TCPEndpointInfoPtr::dynamicCast(p);
You can configure a password prompt callback using the IceSSL plug-in interface see https://doc.zeroc.com/ice/3.7/ice-plugins/icessl/advanced-icessl-topics#id-.AdvancedIceSSLTopicsv3.7-ManagingCertificatePasswords
Adaptive load-balancing is based on the load average (or CPU usage on Windows) of the 2 machines. If the load average is similar on both machines, IceGrid will evenly distribute the requests on the 2 machines.
An easy way to achieve what you want would be to deploy more replicas on the machine which is more powerful. For example, 1 replica on the less powerful machine and 3 replica on the most powerful machine. You can then use the Ordered/Random strategy to distribute the load evenly on all the replicas.
Hi Jose David,
If you want to create an Ice server C++, C#, Java, Python or Swift mappings have full server side support and are best suited for the task.
We don't have immediate plans to support a golang mapping right now. This is something we will consider if there are more demands for it. If this is something you would like to sponsor, you can contact us at [email protected]
My team will move to golang。
ICE is good。BUT。。。
so sad to here that...