Is there a way to get a Clients IP in Context of a write?
I want to get the IP of an Client that writes to my Milo-OPCUA-Server, so I can handle these writes differently based on the Clients IP (local Clients should be able to write directly on the Server, whilst other writes should get forwarded to another Server)
Okay, this is not part of any official API right now, so it almost certainly will break in the future, but:
With the OperationContext you get when implementing AttributeManager#write(WriteContext, List<WriteValue>):
context.getSession().ifPresent(session -> {
UaStackServer stackServer = context.getServer().getServer();
if (stackServer instanceof UaTcpStackServer) {
ServerSecureChannel secureChannel = ((UaTcpStackServer) stackServer)
.getSecureChannel(session.getSecureChannelId());
Channel channel = secureChannel.attr(UaTcpStackServer.BoundChannelKey).get();
SocketAddress remoteAddress = channel.remoteAddress();
}
});
I'll have to add some official API to do this, probably something hanging off the Session object.
Related
I am currently using the ServicePartitionResolver to get the http endpoint of another application within my cluster.
var resolver = ServicePartitionResolver.GetDefault();
var partition = await resolver.ResolveAsync(serviceUri, partitionKey ?? ServicePartitionKey.Singleton, CancellationToken.None);
var endpoints = JObject.Parse(partition.GetEndpoint().Address)["Endpoints"];
return endpoints[endpointName].ToString().TrimEnd('/');
This works as expected, however if I redeploy my target application and its port changes on my local dev box, the source application still returns the old endpoint (which is now invalid). Is there a cache somewhere that I can clear? Or is this a bug?
Yes, they are cached. If you know that the partition is no longer valid, or if you receive an error, you can call the resolver.ResolveAsync() that has an overload that takes the earlier ResolvedServicePartition previousRsp, which triggers a refresh.
This api-overload is used in cases where the client knows that the
resolved service partition that it has is no longer valid.
See this article too.
Yes. They are cached. There are 2 solutions to overcome this.
The simplest code change that you need to do is replace var resolver = ServicePartitionResolver.GetDefault(); with var resolver = new ServicePartitionResolver();. This forces the service to create a new ServicePartitionResolver object to every time. Whereas, GetDefault() gets the cached object.
[Recommended] The right way of handling this is to implement a custom CommunicationClientFactory that implements CommunicationClientFactoryBase. And then initialize a ServicePartitionClient and call InvokeWithRetryAsync. It is documented clearly in Service Communication in the Communication clients and factories section.
My TCP server uses netty.The situation is: When a client connects to the server,I will save the client's ip in a global variable(such as a Map); When the client is disconnected,I will remove the IP from the map.
I used channelConnected() and channelDisconnected() method in SimpleChannelHandler.But my problem is ,some times the channelDisconnected() method cannot catch the event when I think the client is disconnected(maybe the computer closed,or the client process closed,or some other situations...) Can you give me some suggestions.
Just use DefaultChannelGroup which will automatically remove the Channel from it when it was closed.
Alternative you can register a ChannelFutureListener to the Channels close future to do the removal from your map.
Something like this:
channel.getCloseFuture().addListener(new ChannelFutureListener() {
public void operationCompleted(ChannelFuture f) {
map.remove(f.getChannel());
}
});
Because of shared hosting, my redis server on the target host does not run on a port, but on a very specific socket, which can be connected to via the socket file, only accessible to my user.
However, I have not found how I can specify connection via a socket in the node_redis and connect-redis packages, the ones I want to use.
Anyone know how to do it?
Update: My answer below is not really correct. It turns out that the solution in the issue I mention below actually still works. It's more of a coincidence, IMO, but you can do something like this, and it should work:
var redis = require('redis'),
client = redis.createClient('/tmp/redis.sock');
As you see from the code snippet below, this will get passed to net.createConnection which will connect to the unix socket /tmp/redis.sock.
Old answer:
There is a closed issue about this node_redis/issues/204. It seems, thought, that the underlying node.js net.createConnection API has since changed. It looks as though it would be a quite small fix in node_redis' exports.createClient function:
exports.createClient = function (port_arg, host_arg, options) {
var port = port_arg || default_port,
host = host_arg || default_host,
redis_client, net_client;
net_client = net.createConnection(port, host);
redis_client = new RedisClient(net_client, options);
redis_client.port = port;
redis_client.host = host;
return redis_client;
};
It seems as though net.createConnection will attempt to connect to a unix socket if it's called with one argument, that looks like a path. I suggest you implement a fix and send a pull request, since this seems like something worth supporting.
There is no longer a connect string...
var client = redis.createClient(9000); // Open a port on localhost
var client = redis.createClient('/tmp/redis.sock'); // Open a unix socket
var client = redis.createClient(9000, 'example.com');
This, and options are documented on the README.
I am trying to discover items that a pubsub service provides. When I log into the target server, I can get the response successfully. But when I connect bu do not login, it gives a bad request error.
This is the code:
ConnectionConfiguration config = new ConnectionConfiguration(serverAddress, 5222);
config.setServiceName(serviceName);
connection = new XMPPConnection(config);
connection.connect();
connection.login(userName, password); //!!!when I remove this line, bad request error is received
ServiceDiscoveryManager discoManager = ServiceDiscoveryManager.getInstanceFor(connection);
DiscoverItems items;
try {
items = discoManager.discoverItems("pubsubservice." + serverName);
} catch (XMPPException e) {
e.printStackTrace();
}
Is there a way to discover items when the user is not logged in, but the connection is established?
No, you must authenticate to send stanzas to any JID in XMPP (otherwise they would not be able to reply to you, since they wouldn't know your address).
Perhaps one option for you is anonymous authentication. Most servers support it, and it creates a temporary account on the server for you, with a temporary JID. You don't need a password, and login time is quick.
#MattJ is correct and you could try using anon login. That will get you part way there.
Your current request will only get you the nodes though, after which you will have to get the items for each node. It would be simpler to use PubsubManager to get the information you want since it provides convenience methods for accessing/using all things pubsub.
Try the documentation here, the getAffiliations() method is what you are looking for.
BTW, I believe the typical default service name for pubsub is pubsub not pubsubservice. At least this is the case for Openfire.
I know the question isn't very well. Sorry my english.
I want to setup a (one instance of) FreeRadius server to listen to several ports (with a bunch of 'listen' sections) and then pass the that udp port as a parameter along with User-Name and User-Password to a script that I want to use to make the authentication.
The basic idea is make some kind of domain separation. Some Firewall use radius port 2000 to make authentication. Some other different firewall (with a different set of users) use radius port 2020, for example. At the end, all the request fall in the same script that has the knowledge of both set of users and use one or the other according to the given extra attribute (port number)
I know that is possible making a virtual server per 'domain'. but I prefer not to replicate configuration files. and i think is shorter to add a little 'listen' section for every domain I want.
I tried to add an atribute this way:
listen {
ipaddr = *
port = 0
type = auth
update control {
Login-TCP-Port = 1812
}
}
and tried to read it:
autorize {
if ("%{User-Name}" == "bob") {
update reply {
Reply-Message = "This is only %{Login-TCP-Port} an example."
}
update control {
Cleartext-Password := "bob"
}
ok
}
[...]
}
But don't work.
How can i make it right?
Is this posible?
Hope you can help me.
I'm answering myself. I found (looking a like further on google) that the Packet-Dst-Port attribute have the data that I want.
I get it from here (now that I found it, look pretty obvious :P)