Well, I don’t mean to just state the obvious, but the general idea is to push this complexity into the services managed by Amazon. So on the frontend, you would use Amazon Elastic Load Balancing (ELB) to provide highly available load balancing. On the rear end, you use Amazon Relational Database Service (hosted MySQL), SimpleDB, … Read more
The SignalR team has now implemented support for a custom connection factory with StackExchange.Redis, the successor to BookSleeve, which supports redundant Redis connections via ConnectionMultiplexer. The initial problem encountered was that in spite of creating my own extension methods in BookSleeve to accept a collection of servers, fail-over was not possible. Now, with the evolution … Read more
Sounds like you may be looking for a Global Server Load Balancing (GSLB) solution. GSLB usually uses “smart” DNS to direct users to different servers based on a number of parameters (i.e. server unavailable, high load, ip geolocation, etc). As an example, say you have two web servers, one out west (10.10.10.1/24) and one east … Read more
To do this properly, you need to have: Two seperate instances in two datacenters (as you’ve already determined) Synchronisation between the two datacenters (as you’ve already determined) A way of re-directing clients from one to the other in the event of a failure There are two common ways of doing this. One simple, one… not. … Read more
A transaction is a unit of work that you want to treat as “a whole.” It has to either happen in full or not at all. A classical example is transferring money from one bank account to another. To do that you have first to withdraw the amount from the source account, and then deposit … Read more
Typically load balancers are clustered together into a high-availability pair. If one load balancer fails, the secondary picks up the failure and becomes active. They have a heartbeat link between them that monitors status. If all load balancers fail (or are accidentally misconfigured), servers down-stream are knocked offline until the problem is resolved, or you … Read more
If you have so much load that you need to load balance across two haproxy instances then DNS round robin isn’t a bad idea (I would be surprised if you have this load though). DNS round robin won’t provide good failover though. At Stack Overflow we use heartbeat to provide a single virtual IP, this … Read more
Each browser has it’s own method of handling round-robin DNS, I’ve spent some time today researching this problem and will continue to update my answer as I find proof of implementation which will limit my answers to browsers that expose their behavior. Google Chrome Google Chrome (v58 used) will request all host entries for an … Read more
What you want is an active+passive setup. Here’s an example nginx conf snippet to get you going: upstream backend { server 1.2.3.4:80 fail_timeout=5s max_fails=3; server 4.5.6.7:80 backup; } server { listen 80; server_name whatevs.com; location / { proxy_pass http://backend; } } So, ‘normally’, all requests will go to host 1.2.3.4. If we get three failures … Read more
The most important thing that differentiates the two solutions (LVS, HAproxy) is that one is working at layer 4 (LVS) and the other at layer 7 (HAproxy). Note that the layers references are from OSI networking model. If you understand this, you’ll be able to use one in the right place. For example : if … Read more