Reuse gRPC connection/channel when multiple names resolve to the same IPs - grpc

Is there a way to configure/setup multiple gRPC clients to share connections/channels if two or more names/targets resolve to the same remote address (or addresses)?
For example, say I was using the dns name resolver, if dns://service-a and dns://service-b both resolve to [12.34.56.78] I'd like gRPC to know this and decide to only open a single connection to the backend 12.34.56.78.

Related

Correct way to get a gRPC client to communicate with one of many ECS instances of the gRPC service?

I have a gRPC client, not dockerised, and server application, which I want to dockerise.
What I don't understand is that gRPC first creates a connection with a server, which involves a handshake. So, if I want to deploy the dockerised server on ECS with multiple instances, then how will the client switch from one to the other (e.g., if one gRPC server falls over).
I know AWS loadbalancer now works with HTTP 2, but I can't find information on how to handle the fact that the server might change after the client has already opened a connection to another one.
What is involved?
You don't necessarily need an in-line load balancer for this. By using a Round Robin client-side load balancing policy along with a DNS record that points to multiple backend instances, you should be able to get some level of redundancy.

How to encrypt gRPC connections without certificates?

I'm going to be using gRPC for a device to device connection over a network (my device will be running Linux and collecting patient data from various monitors, gRPC will be used by a Windows client system to grab and display that data).
I obviously want to encrypt the data on the wire, but dealing with certificates is going to be a problem for various reasons. I can easily have the server not ask for the client cert, but so far I've been unable to find a way around the client validating the server's cert.
I've got several reasons I don't want to bother with a server cert:
The data collection device (the gRPC server) is going to be assigned an IP and name via DHCP in most cases. Which means that when that name changes (at install time, or when they move the device to a different part of the hospital), I have to automatically fixup the certs. Other than shipping a self-signed CA cert and key with the device, I don't know how to do that.
There are situations where we're going to want to point client to server via IP, not name. Given that gRPC can't do a cert for an IP (https://github.com/grpc/grpc/issues/2691), this becomes a configuration that we can't support without doing something to give a name to a thing we only have an IP for (hosts file on the Windows client?). Given the realities of operating in a hospital IT environment, NOT supporting use of IPs instead of names is NOT an option.
Is there some simple way to accommodate this situation? I'm far from an expert on any of this, so it's entirely possible I've missed something very basic.
Is there some simple way to set the name that the client uses to check the server to be different than the name it uses to connect to the server? That way I could just set a fixed name, use that all the time and be fine.
Is there some way to get a gRPC client to not check the server certificate? (I already have the server setup to ignore the client cert).
Is there some other way to get gRPC to encrypt the connection?
I could conceivably set things up to have the client open an ssh tunnel to the server and then run an insecure gRPC connection across that tunnel, but obviously adding another layer to opening the connection is a pain in the neck, and I'm not at all sure how comfortable the client team is going to be with that.
Thanks for raising this question! Please see my inline replies below:
I obviously want to encrypt the data on the wire, but dealing with
certificates is going to be a problem for various reasons. I can
easily have the server not ask for the client cert, but so far I've
been unable to find a way around the client validating the server's
cert.
There are actually two types of checks happening on the client side: certificate check and the hostname verification check. The former checks the server certificate, to make sure it is trusted by the client; the latter checks the target name with server's identity on the peer certificate. It seems you are suffering with the latter - just want to make sure because you will need to get both of these checks right on the client side, in order to establish a good connection.
The data collection device (the gRPC server) is going to be assigned
an IP and name via DHCP in most cases. Which means that when that name
changes (at install time, or when they move the device to a different
part of the hospital), I have to automatically fixup the certs. Other
than shipping a self-signed CA cert and key with the device, I don't
know how to do that.
There are situations where we're going to want to point client to
server via IP, not name. Given that gRPC can't do a cert for an IP
(https://github.com/grpc/grpc/issues/2691), this becomes a
configuration that we can't support without doing something to give a
name to a thing we only have an IP for (hosts file on the Windows
client?). Given the realities of operating in a hospital IT
environment, NOT supporting use of IPs instead of names is NOT an
option.
gRPC supports IP address(it is also mentioned in the last comment of the issue you brought up). You will have to put your IP address in the SAN field of server's certificate, instead of the CN field. It's true that it will be a problem if your IP will change dynamically - that's why we need DNS domain name, and set up the PKI infrastructure. If that's a bit heavy amount of work for your team, see below :)
Is there some simple way to accommodate this situation? I'm far from
an expert on any of this, so it's entirely possible I've missed
something very basic.
Is there some simple way to set the name that the client uses to check
the server to be different than the name it uses to connect to the
server? That way I could just set a fixed name, use that all the time
and be fine.
You can directly use IP address to connect, and override the target name in the channel args. Note that the overridden name should match the certificate sent from the server. Depending on which credential type you use, it could be slightly different. I suggest you read this question.
Is there some way to get a gRPC client to not check the server
certificate? (I already have the server setup to ignore the client
cert).
Is there some other way to get gRPC to encrypt the connection?
Note that: Even if you don't use any certificate on the wire, if you are sure the correct credential type(either SSL or TLS) is used, then the data on the wire is encrypted. Certificate helps you to make sure the endpoint to which you are connecting is verified. Failing to use certificates will leave your application to Man-In-The-Middle attacks. Hope this can help you better understand the goals and make the right judgement for your team.

Filter out certain grpc exceptions via Envoy?

I'd like to create a reverse proxy to expose several grpc backend services on one host. But I'd also like to whitelist certain grpc status exception categories, and drop all others. I think I've read somewhere that grpc exceptions go into http/2 trailers, so that might be an option.
I'm trying to find info on the grpc wire protocol for passing exceptions, but can't find anything amid all the info on protobuf itself.
Any hints/links ?

Host Name is sometimes empty

In one of my applications (ASP.NET/VB.NET), I need to read the Client Machine Name. Based on the Client Machine we trigger a Point of Sale payment device to accept the payment. On each of these systems we have a stand alone software installed which communicates to the bank using HTTP requests. I am using the following .NET code to read the Client Computer Name.
Dim name As String = String.Empty
Dim hostEntry = Dns.GetHostEntry(HttpContext.Current.Request.UserHostAddress)
If hostEntry.HostName.Contains(".") Then
name = hostEntry.HostName.Substring(0, hostEntry.HostName.IndexOf("."))
Else
name = hostEntry.HostName.Trim
End If
In the development environment, all our systems are in a domain ("xyz.com") and we don't have any issues. In the customer location they don't have a domain name setup. My above logic works well in some of the systems in the client environment and is able to make payments but in most of the systems our logic fails and is not able to read the host name. Any help will be appreciated.
Your question doesn't have the specifics required to answer your question. There are many questions that need to be answered about both environments to give a correct answer. Since I can't ask questions, I will make some assumptions which might apply to future readers of this post and be able to help them out.
I would ask a question but my profile was forked for some unknown reason and I don't have the required reputation to ask a question. That being said I will run through the list of issues I can identify off the bat and suggest solutions for the issue and hopefully one will lead you to a solution.
So...
1) You state you need to read the client machine name. However, if your application isn't running on an internal LAN (aka an intranet) you can't read client machine names period. So this could be your first problem.
2) Combining point 1 and given that you are reading the IP Address from UserHostAddress of the client to look up a DNS host name and when the host look up succeeds you are taking the first part of the name up until the first "." it should be safe to assume that this an intranet application running on a LAN in both your development environment and at the client environment. With that assumption and given the statement that all machines are given an domain of xyz.com it can be assumed that DNS in your development environment is being dynamically updated from presumably through Active Directory (AD). In such case, whenever a client machine on development network requests an IP address, presumably through AD, the DHCP server integrated with AD issues the new IP Address. When it does and the DHCP offer is acknowledged and accepted by the client AD updates DNS (which on a windows network is also AD integrated) by adding a host entry with the computer name of the client machine pointing to the IP Address. Additionally a DNS pointer, depending on configuration, can be added to AD's DNS which allows an IP Address lookup to resolve to the record (which in this case would be the Client's machine name). So with your development environment (presumably running on Windows Active Directory Domain) everything works. Addi tonally, by default the top level domain name (XYZ.COM) gets appended to the clients computer name in initial DNS requests from the client.
3) Your client is not running a domain which leaves further questions. Are they running windows? If they are running windows is it as a non-Ad environment, for example a work group. First assumption would be they are not AD integrated or otherwise you most likely wouldn't be having this problem although I can think of a few rare case scenarios where they might. However, odds are the relevant questions are What DNS server are they running and what DHCP sever are they running? Your application is trying to use a client IP Address on their network and the host name lookup based on their IP is failing so it tells me in their environment for one reason or another you can't get a host name from the IP Address of the client. Mind you if they could be on AD and configured entirely correctly their DNS server is just overwhelmed and not responding within 2 seconds causing the name lookup failure but that is the rate case. With more information I could help more.
3) Assuming in 2 that they are not on AD, do you have the ability manually code host names on the computer your application is running? For example, lets say yourapp.exe runs client-server-01 and clients connect to it. Then on client-server-01 you could add static DNS entries in the host file for each PC on the client network that you expect to connect. On the other hand if your application is running locally on the client PCs you could pass the machine name as a header in the web request and then read it from the Request.Headers variable on the server.
4) Again, making another assumption the clients are web based and your application in the client environment is being hosted on the server... Is the server on a DMZ outside the client environment? If so the client environment may likely be configured, per best practices, that the server host your web app is in a DMZ and DNS requests to the box are forwarded to the client's ISP and not back into their network that has the DNS server capable of resolving an internal IP to a client machine name. If this is the case you need to send the client machine name as a variable from your client or code local IPs to host names in the servers host file (assuming the internal network isn't behind NAT and exposes the real client machine's IP) or request that the DMZ'd server can access the internal DNS and configure the access accordingly.
....
The list really goes on and on but I think I highlighted the problems for 99% of the situations and provided answers to their various solutions.
You can try to take it from X-Forwarded-For header
The X-Forwarded-For (XFF) HTTP header field is a common method for
identifying the originating IP address of a client connecting to a web
server through an HTTP proxy or load balancer.
This is what X-Forwarded-For should return:
X-Forwarded-For: client, proxy1, proxy2
Here some example code:
string ip = Request.ServerVariables["HTTP_X_FORWARDED_FOR"] ;
if (!string.IsNullOrEmpty(ip))
{
string[] ipRange = ip.Split(',');
ip = ipRange[0];
}
else
{
ip = Request.ServerVariables["REMOTE_ADDR"];
}
There was an issue with Firewall setup on the client machine.Due to that our .NET code was failing. After adding an exception to all the incoming requests from xyz.com. My code is working without any issues.
Thank you guys #Alexander Higgins, #halfer for the help.

Standard way of using a single port for multiple sockets?

Hey I am writing an app in Twisted, and as it stands I have 4 servers bound two different ports all communicating with the client via JSON. Is there anyway to bind these 4 servers to the same port and have the interactions remain the same?
For instance say the client subscribes to two different feeds, transmitted via a direct socket.
Right now I just do like
server1.read_string()
server2.read_string()
and it will read the correct JSON string from the respective feeds. Is there anyway to maintain this type of functionality but contact my server on the same port?
I do not want to throw all of the server functionality into one massive server and partition the data by header prefixes.
I don't want to do something like
s = server.read_string()
header = s.split(//some delimiter)[0]
if (header == "SERVER1")
{
// Blahh
}
It sounds like you have many clients interacting with your servers via HTTP. The standard solution is to throw a reverse proxy between the client and your servers - that proxy then forwards connections to the appropriate server depending on the URL. The reverse proxy can run on any one of your existing servers or on its own server to lighten the load.
If your data is cachable, the reverse proxy can do caching on your results too.
There are many reverse proxies available and you will want to choose one based on what sort of workload you have. Do you need it to be highly configurable? Is the data public or based on logins? How long does each connection last / how many connections to you want to hold open at once?
Squid, Varnish, HAProxy are good reverse proxies and even Apache could do this for you.
I plan to use HAProxy for Gridspy, my project as I have many ongoing connections with my clients and want to place an orbited server in the same URL path as my django server. See This tutorial for more information on how to forward many connections on port 80 from one server to many. This tutorial is focused on Comet, but your problem is even simpler than that.
If you are considering an ongoing tcp/ip connection from the browser back to your servers, seriously consider Orbited. See this tutorial about graphs via orbited and morbidQ. Orbited will also punch through firewalls and proxies better than most custom solutions will, as it looks like normal HTTP traffic.
In order to have multiple servers running on the same machine all bound to the same port, they need to be bound to different IP addresses. The only way to bind to the same port on the same IP is to enable the socket's SO_REUSESOCKET option, but then multiple servers would be able to receive each other's inbound data, really messing up your communications.
Otherwise, having a single server that uses headers to identifies the particular feeds is best. Why do you not want to do that?

Resources