Recent content by kemeris

Hello, I’m trying to get connectivity between a Proxmox node and a VM located on the same node but in a tenant VRF (vrf_z10001) proxmox (10.0.4.4 vrf Main) -> VM (10.0.20.10 vrf vrf_z10001) I leaked host routes between VRFs. ip route add 10.0.20.10/32 dev v100001 ip route add 10.0.4.4/32 dev...

Problem solved.

Hello, On Proxmox v8, EVPN Type-5 routes were working fine, but I was not using the Proxmox node as an exit node. After upgrading to Proxmox v9, the nodes no longer import EVPN Type-5 default routes into the VRFs. Whether the Type-5 route is advertised by a BGP peer or the Proxmox node itself...

Hi, When removing a VM via the GUI or API, the VM is not removed from HA Node Affinity Rules. Later, when I try to modify HA Node Affinity Rules via the API by retrieving the current resource list and adding a new resource, I get an error: cannot use unmanaged resource(s) vm:138, vm:169. This...

Hi @Robert Dahlem, I have actually tried /cluster/ha/rules before. Looks like pvesh get /cluster/ha/rules does not return resources. After your examples, I have tried REST API instead, and it does work. Thank you! root@proxmox1-1:~# pvesh get /cluster/ha/rules ┌───────────────────────────────┐...

Hi, to update the rule, "resources" as list should be provided for PUT to /cluster/ha/rules/{rule} . But I don't see how to get current resource list of the rule. GET /api2/json/cluster/ha/rules/{rule} does not allow to get one, so there is no way to append new resource to existing list.

Thanks @ggoller for the reply. That’s helpful — I’ll look for a workaround.

Hello, I am trying to override default rule route-map MAP_VTEP_OUT permit 1 by merging my frr.conf.local frr.conf.local: ip prefix-list C100001_ONLY_IPV4 seq 5 permit 10.0.20.0/24 le 32 ip prefix-list C100001_ONLY_IPV4 seq 10 deny 0.0.0.0/0 le 32 ipv6 prefix-list C100001_ONLY_IPV6 seq 5 permit...

Ah, I completely misunderstood the purpose of "exit nodes," but now it's clear. However, I still can't reach the VMs from the Arista switch. I think my problem is related to the MTU. It looks like the Arista is forcing an MTU of 1330 somewhere, even though the underlay interfaces have their MTU...

Thank you @aderumier, that was very helpful. My setup is Symmetric IRB without MLAG. After some digging I found that Arista wasn’t advertising Type-5 default routes and Type-2 routes for the SVIs. Now it does, but I still can’t ping the VMs. Also, Arista is not listed as an “Exit node” for EVPN...

At the moment, I do not have the Arista as an exit node; I haven’t figured out how to advertise the default route from the Arista yet. Should it be Type-5 route for each VRF? I already have aristaip (192.168.10.1) in vpn controller peer, along with static route to reach it. evpn: evpn1...

Yes, firewall is disabled on proxmox node and VM. Arista does receive Type-2, Type-3 and Type-5 routes from proxmox node (10.0.4.1). Everything looks correct to me. imet routes: arista7060cx(config-router-bgp-vrf-10007)#show bgp neighbors 10.0.4.1 evpn received-routes route-type imet...

Arista: VRF routing table arista7060cx#show ip route vrf all VRF: default Source Codes: C - connected, S - static, K - kernel, O - OSPF, O IA - OSPF inter area, O E1 - OSPF external type 1, O E2 - OSPF external type 2, O N1 - OSPF NSSA external type 1, O N2 - OSPF...

Arista: received routes from Proxmox: arista7060cx#show bgp neighbors 10.0.4.1 evpn received-routes detail BGP routing table information for VRF default Router identifier 192.168.10.1, local AS number 65000 BGP routing table entry for mac-ip 0aa9.2c3c.b7a0, Route Distinguisher: 10.0.4.1:8...

Hello everyone, I am trying to connect my 3 node Proxmox cluster to Arista 7060CX switch via iBGP so Arista could act as "Exit node" for Proxmox SDN Zones. I am new to Arista. I have managed to establish iBGP session with Arista, map evpn routes to corresponding VRF's and VLANS to mimic network...