Not sure if this has been addressed elsewhere but I attempted to search on the classic load balancer. I have been informed by AWS that as of August 15th classic load balancer will no longer be supported as it is running on the EC2 classic network which is being depreciated. Sounds like you must convert to an application load balancer in ELB. I was working on this last night and ran into some connectivity issues where it can't determine the health of the instance. I'm going to work with AWS support to get this resolved but I thought others should be aware this could be an impending problem for those running classic load balancers. If I have misunderstood this end-of-life requirement from AWS please let me know
Hi @Slickrock22
Yes, ALB is the only thing that has been available for a year I think. And ELB is being removed completely.
Perhaps there are other things in your setup, which is "older"? Have you seen this, I know you're using EC2 not BS, but it should be similar. We do not have a specific guide for ALB by itself:
https://www.metabase.com/docs/latest/operations-guide/running-metabase-on-elastic-beanstalk.html#step-2---configure-the-basic-metabase-architecture
Yes I have read through that extensively. Actually I am using Beanstalk. I think in the process you have to actually set the networking configuration to use the VPC before you are able to select the newer application load balancer. Otherwise it will give me an error when I try. Last night I attempted to create a brand new instance of metabase on Beanstalk from scratch using ALB and it is giving me an error about the health of the instance. I will troubleshoot today and see what the issue is I just wanted to make sure that everyone was aware that instances using the classic load balancer may not work any longer after August 15. I did also noticed that there is a upgrade option in the classic load balancer But I don’t know enough to know if that is a good idea and would prefer just to use a standard beanstalk approach so I can replicate.
@Slickrock22 You're definitely stuck in a hard place, since if the auto-upgrade from ELB to ALB fails or causes problems, then I don't think you have a way to revert. So the best option would be to go with all new, once that's working, then you can copy the application database data.
@flamber I was successfully able to complete this transition. I'm not sure if I was doing something wrong but I was not able to follow the Metabase documentation sequentially for launching in AWS beanstalk. You are not able to switch to an application load balancer during setup if you do not first specify that you want to launch in a VPC. In the Metabase documentation, VPC configuration is after load balancer selection equals application. Also I needed to add HTTP code equal 200 for the health check under processes and load balancer. Without that my instances would fail to start as they failed the initial health check. I'm not sure if this last specification for HTTP code was required or not but AWS tech support said they SSH onto the instance it was looking for 200. I added this in the initial setup configuration and it worked. Hopefully, this information helps somebody else and potentially can be the basis to update the documentation if what I am sharing is accurate.