@matteo-nunziati said in how do you deal with SPOF with HAproxy: @stacksofplates said in how do you deal with SPOF with HAproxy: Keepalived is most likely what you're looking for. You assign a VIP to your interface and it keeps a heartbeat between the systems. The VIP (floating IP) will move between systems if there is an issue. Once that's set up, use Serf to update your HAProxy configs or Consul for automatic service discovery which HAProxy can read. If you're using a cloud provider, I'd just use one of their provided load balancers. +1 for serf but I can not find any doc about serf as a distributed config (as consul). I know about serf as a discovery/alive tool only... When systems send their messages to the cluster they can run a script depending on the message. Their example repository has simple Bash scripts to update an HAProxy config. https://github.com/hashicorp/serf/blob/master/demo/web-load-balancer/README.md It's very very simple, but sometimes that's easier than setting up a full Consul cluster.

@Dashrender said in Moving to Cloudflare proxy: Seriously you want me to buy another domain to fix this? Yes, because someone screwed up with the original domain, so yes, you need to either fix that or do something to work around it. Mistakes have costs, this is a pretty trivial one.

@bradhawkins said in My config for HAPRoxy and Exchange 2013: Thank you so much for sharing your config, I have not found anything else that works for Exchange 2016 and this works perfectly. Glad it works for you. I do not have a client with on prem 2016 to test with.

https://serverfault.com/questions/663855/haproxy-for-load-balancing-vsftpd-servers

@jaredbusch said in HAProxy fails to start on reboot: @scottalanmiller said in HAProxy fails to start on reboot: In /etc/sysctl.conf can you add this line? net.ipv4.ip_nonlocal_bind = 1 Or just do... echo "net.ipv4.ip_nonlocal_bind = 1" >> /etc/sysctl.conf That did it. WTF did I just do? I can read, so I get the general idea. For load balancing it needs to be able to bind to non-local ports. For some reason the install did not set this which it seems like it should have done.

@jaredbusch said in HAProxy not workign with SELinux enforcing: Apparently i never setup this server to auto update... Ayup..... [[email protected] ~]# dnf -y update Last metadata expiration check: 0:08:45 ago on Mon 27 Nov 2017 06:44:36 PM CST. Dependencies resolved. ================================================================================================================================== Package Arch Version Repository Size ================================================================================================================================== <snip> Upgrading: selinux-policy noarch 3.13.1-260.14.fc26 updates 508 k selinux-policy-targeted noarch 3.13.1-260.14.fc26 updates 9.4 M <snip> Transaction Summary ================================================================================================================================== Install 3 Packages Upgrade 43 Packages Remove 3 Packages

@jaredbusch said in How can I find HAProxy log history with Fedora 26: @stacksofplates well that was a bust. it only shows errors related to the service itself. journalctl -u haproxy I need to find connection errors. Will hit the googles in an hour when I arrive on site. Ah ok, I didn't realize it doesn't log that automatically ( I don't ever use HAProxy). This may help from https://serverfault.com/questions/647395/haproxy-not-logging-requests

@Ambarishrh said: was wondering the same, they could move all the Linux and save quite a lot of cost They are "all in" on MS technologies. I followed them when they were building the system. Their sponsor is a 100% MS devotee (he's the father of VBA, for that matter) and there is no way that they would consider something based on logic. They were the pioneer user of the .NET MVC system and everything they have done is based on total lock in to MS, which has its advantages. But overall, they are using costly, slow components to do work. I'm sure that it works pretty well, but as good as it could? No way. We've seen other communities like that make odd technology decisions leaving them locked in to old schemes and costing a fortune to do what is cheap with modern design choices.

@JaredBusch said: @coliver said: @JaredBusch said: Disabke SELinux and see if it works. Then if so, add the port as allowed to SELinux? Looks like it was SELinux, setting it to permissive allowed the port to be used. Now to figure out how to allow the port in SELinux. semanage port -m -t http_port_t -p tcp 82 That did it thanks.

HAProxy is very cool. It is used nearly everywhere that I look with cloud hosted applications.