Bug 2232279
| Summary: | Add firewall rules to allow ovn traffic from dcn site | ||
|---|---|---|---|
| Product: | Red Hat OpenStack | Reporter: | Jaison Raju <jraju> |
| Component: | openstack-tripleo-heat-templates | Assignee: | OSP Team <rhos-maint> |
| Status: | NEW --- | QA Contact: | Joe H. Rahme <jhakimra> |
| Severity: | unspecified | Docs Contact: | |
| Priority: | unspecified | ||
| Version: | 17.1 (Wallaby) | CC: | jraju, mburns, tkajinam |
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | x86_64 | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | If docs needed, set a value | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | Type: | Bug | |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
Description of problem: It was observed that the controller nodes had allow rule for traffic to 6642 from internal_api network from its leaf only and not from other leafs or dcn sites. For example: We have 2 sites central & dcn1 with 172.18.1.0/24 & 172.18.2.0/24, controller only allowed this: [root@central-controller0-0 ~]# iptables-save | grep 6642 -A INPUT -s 172.18.1.0/24 -p tcp -m tcp --dport 6642 -m conntrack --ctstate NEW -m comment --comment "121 OVN DB server and cluster ports for 172.18.1.0/24 ipv4" -j ACCEPT This caused the computes to not be able to get the required ovn wiring like tunnels or port bind post deployment. Version-Release number of selected component (if applicable): 17.1 / RHOS-17.1-RHEL-9-20230628.n.2 How reproducible: Always Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info: I feel like we need to consider adding firewall rules similar to this for each dcn site: -A INPUT -s 172.18.1.0/24 -p tcp -m tcp --dport 6644 -m conntrack --ctstate NEW -m comment --comment "121 OVN DB server and cluster ports for 172.18.1.0/24 ipv4" -j ACCEPT -A INPUT -s 172.18.1.0/24 -p tcp -m tcp --dport 6643 -m conntrack --ctstate NEW -m comment --comment "121 OVN DB server and cluster ports for 172.18.1.0/24 ipv4" -j ACCEPT -A INPUT -s 172.18.1.0/24 -p tcp -m tcp --dport 6642 -m conntrack --ctstate NEW -m comment --comment "121 OVN DB server and cluster ports for 172.18.1.0/24 ipv4" -j ACCEPT I guess the best way to do this would be to make sure the final deployment on central site after adding any dcn site should consider separate net_cidr for each site. (Looking at the file, I am not sure whether it was supposed to apply for each site, it in my environment the rules for dcn site were missing): /usr/share/openstack-tripleo-heat-templates/deployment/ovn/ovn-dbs-cluster-ansible.yaml role_data: description: Role data for the OVN multi-active cluster role. value: service_name: ovn_dbs firewall_rules: map_merge: repeat: for_each: <%net_cidr%>: get_param: - ServiceData - net_cidr_map - {get_param: [ServiceNetMap, OvnDbsNetwork]} template: '121 OVN DB server and cluster ports for <%net_cidr%>': proto: 'tcp' source: <%net_cidr%> dport: - {get_param: OVNNorthboundServerPort} - {get_param: OVNSouthboundServerPort} - {get_param: OVNNorthboundClusterPort} - {get_param: OVNSouthboundClusterPort}