Bug 1605218
| Summary: | [Netvirt] Overlay Tunnel not created between controller and compute node | ||||||
|---|---|---|---|---|---|---|---|
| Product: | Red Hat OpenStack | Reporter: | Janki <jchhatba> | ||||
| Component: | opendaylight | Assignee: | Tim Rozet <trozet> | ||||
| Status: | CLOSED WONTFIX | QA Contact: | Noam Manos <nmanos> | ||||
| Severity: | high | Docs Contact: | |||||
| Priority: | medium | ||||||
| Version: | 13.0 (Queens) | CC: | aadam, jchhatba, mkolesni, trozet, vthapar | ||||
| Target Milestone: | z5 | Keywords: | Triaged, ZStream | ||||
| Target Release: | 13.0 (Queens) | ||||||
| Hardware: | Unspecified | ||||||
| OS: | Unspecified | ||||||
| Whiteboard: | Netvirt | ||||||
| Fixed In Version: | Doc Type: | If docs needed, set a value | |||||
| Doc Text: | Story Points: | --- | |||||
| Clone Of: | Environment: | ||||||
| Last Closed: | 2019-03-06 16:14:53 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: | |||||||
| Bug Depends On: | 1642270 | ||||||
| Bug Blocks: | |||||||
| Attachments: |
|
||||||
This is particularly seen only during update. ITM Direct tunnel is expected to solve this issue. Can we retry the usecase by setting itm-direct-tunnels flag to true in genius-ifm-config.xml (In reply to Aswin Suryanarayanan from comment #11) > ITM Direct tunnel is expected to solve this issue. Can we retry the usecase > by setting itm-direct-tunnels flag to true in genius-ifm-config.xml Direct tunnel may not help with this. From the logs, I can see tunnels present in OVSDB config but not on switch. This looks like OVSDB counterpart of missing default flows issue where switches connect before plugin is fully ready. There were no OVSDB logs in karaf logs for this issue so couldn't confirm the theory. Janki mentioned that we check for netvirt:1 before configuring manager on switch. We should change this check to diagstatus to make sure OVSDB is fully up before we connect switches. As per discussions with janki, I'll raise a bug on puppet-neutron to change the check to diagstatus before setting manager on OVS. *** Bug 1638724 has been marked as a duplicate of this bug. *** This is also happening after update, please see attachment 1493357 [details] from bug 1638724 Tim, I see the patch was merged upstream. Should this move to POST? As per depreciation notice [1], closing this bug. Please reopen if relevant for RHOSP13, as this is the only version shipping ODL. [1] https://access.redhat.com/documentation/en-us/red_hat_openstack_platform/14/html-single/release_notes/index#deprecated_functionality As per depreciation notice [1], closing this bug. Please reopen if relevant for RHOSP13, as this is the only version shipping ODL. [1] https://access.redhat.com/documentation/en-us/red_hat_openstack_platform/14/html-single/release_notes/index#deprecated_functionality |
Created attachment 1464952 [details] karaf logs, data dumps, ovs logs, flows, OS CLI outputs, sosreport Description of problem: An overlay tunnel is not created between controller-1 and compute-1. tun port is present on compute but not on controller node. The port was absent even after spawning a VM. Version-Release number of selected component (if applicable): OSP13 + ODL How reproducible: sometimes Actual results: Expected results: tun port should be created. Additional info: