Bug 2037739
| Summary: | VM remains locked after importing from vmware/external-ova, we see "'str' object has no attribute 'decode'" in the log | ||
|---|---|---|---|
| Product: | Red Hat Enterprise Virtualization Manager | Reporter: | Michal Skrivanek <michal.skrivanek> |
| Component: | vdsm | Assignee: | Milan Zamazal <mzamazal> |
| Status: | CLOSED ERRATA | QA Contact: | Tamir <tamir> |
| Severity: | high | Docs Contact: | |
| Priority: | unspecified | ||
| Version: | 4.4.9 | CC: | ahadas, bugs, dfodor, emarcus, lsurette, nsurati, pagranat, pelauter, srevivo, tamir, tgolembi, ycui, Yury.Panchenko |
| Target Milestone: | ovirt-4.4.9-3 | Keywords: | ZStream |
| Target Release: | 4.4.9 | ||
| Hardware: | x86_64 | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | No Doc Update | |
| Doc Text: |
Previously, in some cases of importing a virtual machine, the VDSM logs were filled with repeated errors regarding a missing attribute. As a result, although the import process succeeded, the virtual machine sometimes remained locked, and the debug process was difficult due to the repeated errors in the VDSM log.
In this release, the import process succeeds, the resulting virtual machine remains unlocked, and the VDSM log is not filled with the missing attribute error.
|
Story Points: | --- |
| Clone Of: | 2026809 | Environment: | |
| Last Closed: | 2022-01-12 14:44:14 UTC | Type: | --- |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | Virt | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
| Bug Depends On: | 2026809 | ||
| Bug Blocks: | |||
|
Description
Michal Skrivanek
2022-01-06 13:28:26 UTC
cherry pick https://gerrit.ovirt.org/#/c/vdsm/+/118105/ Hi Michal, please review this text for the Erratum: "Previously, when importing a virtual machine, VDSM logs were filled with repeated errors regarding a missing attribute. As a result, although the import process succeeded, the virtual machine sometimes remained locked, and the debug process was difficult due to the multiple repeated errors in the VDSM log. In this release, the import process succeeds, the resulting virtual machine remains unlocked, and the VDSM log is not filled with the missing attribute error. " (In reply to Eli Marcus from comment #4) > Hi Michal, please review this text for the Erratum: > > "Previously, when importing a virtual machine, VDSM logs were filled with > repeated errors regarding a missing attribute. As a result, although the > import process succeeded, the virtual machine sometimes remained locked, and > the debug process was difficult due to the multiple repeated errors in the > VDSM log. > In this release, the import process succeeds, the resulting virtual machine > remains unlocked, and the VDSM log is not filled with the missing attribute > error. " Hi Eli, I'm afraid we don't have enough details to say this - we don't know yet of any flow that would have resulted in a valid (successful) import that failed due to the issue that is reflected in that message. We suspect that the error handling was incorrect due to that issue - so instead of failing properly, the VM remained lock. This happened also for us. That said, we reached out to Yury in order to get more details on the failure they experienced and how it was resolved with that fix. Verified on RHV 4.4.9-6 and RHVH 4.4.9 host.
Env:
- Engine instance with RHV 4.4.9-6 (ovirt-engine-4.4.9.5-0.1.el8ev.noarch) and RHEL 8.5 installed.
- 1 host with redhat-virtualization-host-4.4.9-202201072228_8.5 and with vdsm-4.40.90.5-1.el8ev included.
Tests:
Check the partner's case:
I performed the steps from Veeam's user guide (from page 20):
1. I have downloaded the zip from their website, which can be alternatively retrieved from the following Red Hat server.
2. I have extracted it in /tmp/new_folder in one of the hosts in the env.
3. I ran chmod 755 veeam_rhv_proxy_beta_vm_1.0.1488.* and
chown 36:36 veeam_rhv_proxy_beta_vm_1.0.1488.*
In Admin Portal:
4. Create a 4.6 data center and a 4.6 cluster.
5. Install the hosts and create a new NFS storage domain.
6. Open the VM import modal, select OVA as the source, fill in the OVA path and host, load, and import the VM.
7. The VM was imported successfully and was unlocked.
Results (As Expected):
4. The 4.6 data center and the 4.6 cluster were created.
5. The host was installed and the NFS storage domain was created.
6. The VM import process has started.
7. The VM was imported successfully and was unlocked.
Check that the log doesn't contain any mentions of the "'str' object has no attribute 'decode'" error:
In Admin Portal:
1. Create a 4.6 data center and a 4.6 cluster.
2. Install the hosts and create a new NFS storage domain.
3. Open the VM import modal, select VMware as the source, fill in the VCenter 7 details, and select a VM to import and import it.
4. Check that in the engine.log and vdsm.log of the host used in the import there aren't any mentions of the "'str' object has no attribute 'decode'" error while the import process runs.
5. Check that the VM was imported successfully.
Results (As Expected):
1. The 4.6 data center and the 4.6 cluster were created.
2. The host was installed and the NFS storage domain was created.
3. The VM import process has started.
4. In the engine.log and vdsm.log in the host used in the import there aren't any mentions of the "'str' object has no attribute 'decode'" error while and after the import process.
5. The VM was imported successfully.
Check that import from a place with no permissions (root specifically) fails.
1. I have downloaded the zip from their website, which can be alternatively retrieved from the following Red Hat server.
2. I have extracted it in /root in one of the hosts in the env.
3. I ran chmod 755 veeam_rhv_proxy_beta_vm_1.0.1488.* and
chown 36:36 veeam_rhv_proxy_beta_vm_1.0.1488.*
In Admin Portal:
4. Create a 4.6 data center and a 4.6 cluster.
5. Install the hosts and create a new NFS storage domain.
6. Open the VM import modal, select OVA as the source, fill in the OVA path and host, load, and import the VM.
7. Check that VM import has failed and there isn't a locked instance of it.
Results (As Expected):
4. The 4.6 data center and the 4.6 cluster were created.
5. The host was installed and the NFS storage domain was created.
6. The VM import process has started.
7. The VM import has failed and there isn't a locked instance of it.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory (RHV RHEL Host (ovirt-host) [ovirt-4.4.9] ASYNC #2), and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2022:0134 |