This site requires JavaScript to be enabled to function correctly, please enable it.
Note: This bug is displayed in read-only format because
the product is no longer active in Red Hat Bugzilla.
Summary:
Parameter duplication for configuring openvswitch
Product:
Red Hat OpenStack
Reporter:
Miguel Angel Nieto <mnietoji>
Component:
openstack-tripleo Assignee:
Vijayalakshmi Candappa <vcandapp>
Status:
CLOSED
WONTFIX
QA Contact:
Joe H. Rahme <jhakimra>
Severity:
high
Docs Contact:
Priority:
medium
Version:
17.1 (Wallaby) CC:
bshephar, cfontain, ekuris, gregraka, hakhande, jamsmith, mburns, ramishra, sbaker, supadhya, vcandapp
Target Milestone:
--- Keywords:
Triaged
Target Release:
--- Flags:
vcandapp:
needinfo-
needinfo-
Hardware:
Unspecified
OS:
Unspecified
Whiteboard:
Fixed In Version:
Doc Type:
If docs needed, set a value
Doc Text:
Story Points:
---
Clone Of:
Environment:
Last Closed:
2024-02-21 16:38:21 UTC
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: In 16.2 we used to configure ovs in the following way: arameter_defaults: ComputeOvsDpdkSriovR730Parameters: KernelArgs: "default_hugepagesz=1GB hugepagesz=1G hugepages=64 iommu=pt intel_iommu=on tsx=off isolcpus=2-19,22-39" NeutronBridgeMappings: "dpdk-mgmt:br-link0,dpdk-data0:br-dpdk0,dpdk-data1:br-dpdk1" IsolCpusList: "2-19,22-39" NovaComputeCpuDedicatedSet: ["8-19,28-39"] NovaReservedHostMemory: 4096 OvsDpdkSocketMemory: "4096,4096" OvsDpdkMemoryChannels: "4" OvsPmdCoreList: "2,3,4,5,6,7,22,23,24,25,26,27" NovaComputeCpuSharedSet: [0, 20, 1, 21] OvsPmdAutoLb: true OvsPmdLoadThreshold: 70 OvsPmdImprovementThreshold: 25 OvsPmdRebalInterval: 2 In 17.0 this configuration can be done through provisioning in the following way: ansible_playbooks: - playbook: /home/stack/ospd-17.1-geneve-ovn-dpdk-sriov-ctlplane-dataplane-bonding-hybrid-mq/playbooks/bls_workaround.yaml - playbook: /usr/share/ansible/tripleo-playbooks/cli-overcloud-node-kernelargs.yaml extra_vars: kernel_args: "default_hugepagesz=1GB hugepagesz=1G hugepages=64 iommu=pt intel_iommu=on tsx=off isolcpus=2-19,22-39" reboot_wait_timeout: 900 tuned_profile: "cpu-partitioning" tuned_isolated_cores: "2-19,22-39" - playbook: /usr/share/ansible/tripleo-playbooks/cli-overcloud-openvswitch-dpdk.yaml extra_vars: memory_channels: "4" lcore: "0,20,1,21" pmd: "2,3,4,5,6,7,22,23,24,25,26,27" socket_mem: "4096,4096" pmd_auto_lb: true pmd_load_threshold: "70" pmd_improvement_threshold: "25" pmd_rebal_interval: "2" I tried to configure only in the provisioning stage and remove those parameters from de deployment stage, but it is not possible, I get an error: 2022-11-02 08:39:59,728 p=129214 u=stack n=ansible | 2022-11-02 08:39:59.727949 | 525400c8-ce75-8572-41f5-000000000d44 | TASK | Skipping reboot for deployed node 2022-11-02 08:39:59,743 p=129214 u=stack n=ansible | 2022-11-02 08:39:59.742855 | 525400c8-ce75-8572-41f5-000000001409 | TASK | Check valid input for tripleo_ovs_dpdk_pmd_core_list 2022-11-02 08:39:59,753 p=129214 u=stack n=ansible | 2022-11-02 08:39:59.750884 | 525400c8-ce75-8572-41f5-000000000962 | SKIPPED | ddp package selection | compute-r740 2022-11-02 08:39:59,755 p=129214 u=stack n=ansible | 2022-11-02 08:39:59.754699 | 525400c8-ce75-8572-41f5-000000000d44 | SKIPPED | Skipping reboot for deployed node | controller-2 2022-11-02 08:39:59,782 p=129214 u=stack n=ansible | 2022-11-02 08:39:59.782050 | 525400c8-ce75-8572-41f5-000000000964 | TASK | Run ovs-dpdk role 2022-11-02 08:39:59,789 p=129214 u=stack n=ansible | 2022-11-02 08:39:59.788226 | 525400c8-ce75-8572-41f5-000000001409 | FATAL | Check valid input for tripleo_ovs_dpdk_pmd_core_list | compute-r730 | error={"changed": false, "msg": "List of PMD cores cannot be empty - tripleo_ovs_dpdk_pmd_core_list"} 2022-11-02 08:39:59,861 p=129214 u=stack n=ansible | 2022-11-02 08:39:59.859855 | 525400c8-ce75-8572-41f5-000000000040 | SKIPPED | include_tasks | controller-2 2022-11-02 08:39:59,919 p=129214 u=stack n=ansible | 2022-11-02 08:39:59.918613 | 4044822b-ed66-4404-ba09-1dec7a6b84ab | INCLUDED | /usr/share/ansible/roles/tripleo_ovs_dpdk/tasks/config.yml | compute-r740 2022-11-02 08:39:59,945 p=129214 u=stack n=ansible | 2022-11-02 08:39:59.945553 | 525400c8-ce75-8572-41f5-000000001475 | TASK | Check valid input for tripleo_ovs_dpdk_pmd_core_list 2022-11-02 08:39:59,952 p=129214 u=stack n=ansible | 2022-11-02 08:39:59.950882 | 525400c8-ce75-8572-41f5-000000000041 | SKIPPED | include_tasks | controller-2 2022-11-02 08:39:59,978 p=129214 u=stack n=ansible | 2022-11-02 08:39:59.977621 | 525400c8-ce75-8572-41f5-000000001475 | FATAL | Check valid input for tripleo_ovs_dpdk_pmd_core_list | compute-r740 | error={"changed": false, "msg": "List of PMD cores cannot be empty - tripleo_ovs_dpdk_pmd_core_list"} I think that this issue creates confusion about how to configure parameters Version-Release number of selected component (if applicable): RHOS-17.1-RHEL-9-20221004.n.1 I tested in 17.1, but I think this also happen in 17.0 How reproducible: Configure ovs parameter only in the provisioning stage Actual results: Ovs parameters needs to be duplicated in order to have a sucessfull deployment Expected results: Ovs should be defined only in one place Additional info: