#dev | Logs for 2024-12-11
« return
[02:27:10] <kolie> Alright back to this.
[02:27:29] <kolie> I got a vlan created and setup outbound nat on the thing so I can get the hosts updated and running.
[02:30:39] <kolie> Looks like the ports need vlan 3002 if they dont have 4000
[02:31:07] <kolie> so thats GE 10/11/12
[02:36:16] <kolie> booting up the first k8s host
[03:04:15] -!- systemd [systemd!~systemd@50.54.nrl.yto] has joined #dev
[03:04:15] -!- systemd has quit [Changing host]
[03:04:15] -!- systemd [systemd!~systemd@pid1] has joined #dev
[03:10:56] <robc207> Hello Bytram!
[03:13:03] <robc207> kolie please reserve a subnet which we will use for serviceType: LoadBalancer requests. Must be in the same subnet as the hosts
[03:15:13] <robc207> I guess that's misleading. I like to route a /24 for the hosts network and then I create a DHCP reservation for the hosts on the lower half and use the upper half for MetalLB
[03:16:13] <kolie> I've got 172.16.42.0/24 configured right now
[03:16:27] <kolie> .100, .101, and .102 for the hosts.
[03:16:43] <kolie> 100 is setup and adding keys to it rn.
[03:29:47] <kolie> 101 is getting installed rn, 102 host is getting prepped.
[03:42:21] <kolie> 101 done just cleaning up 102 now.
[03:45:43] <kolie> Alright all three hosts prepped and up to date.
[03:54:26] <kolie> root@sn-kube-100:/home/kolie/.ssh# k3s kubectl get nodes
[03:54:26] <kolie> NAME STATUS ROLES AGE VERSION
[03:54:26] <kolie> sn-kube-100 Ready control-plane,etcd,master 106s v1.31.3+k3s1
[03:54:26] <kolie> sn-kube-101 Ready control-plane,etcd,master 2m5s v1.31.3+k3s1
[03:54:26] <kolie> sn-kube-102 Ready control-plane,etcd,master 2m58s v1.31.3+k3s1
[04:59:25] -!- chromas [chromas!~chromas@Soylent/Staph/Infector/chromas] has joined #dev
[04:59:33] -!- systemd has quit [Ping timeout: 272 seconds]
[04:59:33] -!- chromass has quit [Ping timeout: 272 seconds]
[06:33:17] -!- chromas has quit [Ping timeout: 272 seconds]
[06:33:35] -!- chromas [chromas!~chromas@Soylent/Staph/Infector/chromas] has joined #dev
[10:26:59] -!- chromas has quit [Ping timeout: 272 seconds]
[10:27:15] -!- chromas [chromas!~chromas@Soylent/Staph/Infector/chromas] has joined #dev
[14:46:39] -!- chromas has quit [Ping timeout: 272 seconds]
[14:46:55] -!- chromas [chromas!~chromas@Soylent/Staph/Infector/chromas] has joined #dev
[23:23:06] <kolie> alright so I got all the nodes chilling in 172.16.42.0/24 100,101 and 102. I have SSH forwarded to each of the hosts via NAT from the router, I don't have any IPs specifically routing to the cluster otherwise. I have BGP at the spine, but the cluster is just base k3s rn.
[23:24:05] <kolie> Are we aiming to do metallb layer 3 bgp for loadbalancing or did you have something else in mind
[23:25:05] <kolie> Already using traefik in the docker host, and def like traefik as a frontend already.
[23:26:57] <kolie> I guess the other thing is how you suggest handling pvcs
[23:56:52] -!- systemd [systemd!~systemd@50.54.nrl.yto] has joined #dev
[23:56:52] -!- systemd has quit [Changing host]
[23:56:52] -!- systemd [systemd!~systemd@pid1] has joined #dev
[23:57:44] <chromas> Are we switching over to kube?