Bug 907431
| Summary: | 3.0 - Migration from vdsm-4.9-113 to vdsm-4.9.6-44 fails | ||||||
|---|---|---|---|---|---|---|---|
| Product: | Red Hat Enterprise Virtualization Manager | Reporter: | Tomas Dosek <tdosek> | ||||
| Component: | Documentation | Assignee: | Cheryn Tan <chetan> | ||||
| Status: | CLOSED WORKSFORME | QA Contact: | ecs-bugs | ||||
| Severity: | high | Docs Contact: | |||||
| Priority: | high | ||||||
| Version: | 3.0.8 | CC: | abaron, acathrow, asegundo, bazulay, dyasny, hateya, lpeer, michal.skrivanek, mkalinin, rlandman, thildred, ykaul | ||||
| Target Milestone: | --- | Keywords: | Regression, Reopened | ||||
| Target Release: | 3.2.0 | ||||||
| Hardware: | Unspecified | ||||||
| OS: | Unspecified | ||||||
| Whiteboard: | virt | ||||||
| Fixed In Version: | Doc Type: | Bug Fix | |||||
| Doc Text: | Story Points: | --- | |||||
| Clone Of: | Environment: | ||||||
| Last Closed: | 2013-04-06 18:43:31 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: | |||||||
| Attachments: |
|
||||||
|
Description
Tomas Dosek
2013-02-04 11:32:48 UTC
Created attachment 692681 [details]
Host sosreport - from host with vdsm 4.9.6-44
This looks like a FC domain? (SOLR-FC_DataCentre) but the VG is not visible. Assuming that the relevant domain is indeed the one that cannot be found this is probably a configuration issue (lun mapping in storage side not set for the new host?) The lun mapping was set up correctly, the host was only reinstalled with newer RHEV-H iso. When we pushed back the older ISO (did the very same settings) it worked fine. the log is missing any date regarding the migration - we can only see errors/warning accessing the storage. If that's the case the problem is not related to migration. Please retest or confirm The time of issue is 2013-02-04 06:01:50,396 - I was connected to the customer and performing the migration. So we may be sure the time mentioned in Description is correct. No storage issues were to be seen at that time, only change was placing newer vdsm to host. I see no migration issues in the logs, only OS level error that ISO domain can not be mounted, that's possible storage server and/or client setup issue, not VDSM's one. We couldn't reproduce it in ours mixed 3.1/3.0 setups. Theoretically, there is a little chance it could be caused by 3.1 → 3.0 migration logic, but I do not see how. Since the original setup right now can not be used to reproduce the bug, I see no possibility to trace it. As per mailing thread about this issue, I'm changing the bug's component to documentation to document this issue. Content of vdsm documentation should contain note that would warn user that live migration doesn't have to necessarily work in case various versions of vdsm package are deployed on hosts within the same cluster. Example: Cluster C contains host A and host B, host A is equipped with vdsm-4.9-113, host B is equipped with vdsm newer than 4.9.6-44. In this scenario it's expected result that some live migration flows would not work correctly. Hey Tomas, I'm going to change the component of this bug from rhevm-doc (which is the rpm package that ships with RHEVM to provide context sensitive help) to the Guides component, which is where we track issues requiring documentation. That being said, we're working hard on getting the 3.2 documentation ready for the upcoming 3.2 release, and won't be looking at bugs against older versions of the documentation for a while. per testing by marina - vdsm 4.9-113 live migration works with vdsm 4.10.1-9 |