oVirt 4.1 not possible to set local maintenance on single host
Hello, I have a single host test environment on 4.1.0 with hosted engine VM. I'm going to update to 4.1.1. Normally the workflow was: - put env in global maintenance - update engine part with engine-setup and such - update other os related packages of hosted engine VM - shutdown hosted engine vm - exit global maintenance Verify engine VM starts and all is ok form web admin gui. These steps above I have already done and now my engine vm has latest 4.1.1 setup. Now I want to proceed also with the only existing host part and I have already run: - shutdown all running VMs - put env in global maintenance - shutdown hosted engine vm Status is: [root@ractor ~]# hosted-engine --vm-status !! Cluster is in GLOBAL MAINTENANCE mode !! --== Host 1 status ==-- conf_on_shared_storage : True Status up-to-date : True Hostname : ractor.mydomain Host ID : 1 Engine status : {"reason": "bad vm status", "health": "bad", "vm": "down", "detail": "down"} Score : 3400 stopped : False Local maintenance : False crc32 : 017f5635 local_conf_timestamp : 5595857 Host timestamp : 5595835 Extra metadata (valid at timestamp): metadata_parse_version=1 metadata_feature_version=1 timestamp=5595835 (Thu Apr 20 12:06:21 2017) host-id=1 score=3400 vm_conf_refresh_time=5595857 (Thu Apr 20 12:06:44 2017) conf_on_shared_storage=True maintenance=False state=GlobalMaintenance stopped=False !! Cluster is in GLOBAL MAINTENANCE mode !! [root@ractor ~]# Normally I set host into local maintenance now, but I receive this error: [root@ractor ~]# hosted-engine --set-maintenance --mode=local Unable to enter local maintenance mode: there are no available hosts capable of running the engine VM. [root@ractor ~]# [root@ractor ~]# ps -ef|grep [k]vm root 887 2 0 Feb14 ? 00:00:00 [kvm-irqfd-clean] [root@ractor ~]# Is this a bug or changed functionality? Thanks, Gianluca
Hi, global maintenance is meant for completely manual changes where all logic needs to be disabled. That includes maintenance migrations. You should exit global maintenance and put just the single host to local maintenance. Preferably using the webadmin UI (Management / Maintenance). That will migrate all VMs away (including hosted engine if it runs there) and disconnect all storage domains. This might be tricky if all you have is just a single host. Best regards -- Martin Sivak SLA / oVirt On Thu, Apr 20, 2017 at 12:12 PM, Gianluca Cecchi <gianluca.cecchi@gmail.com> wrote:
Hello, I have a single host test environment on 4.1.0 with hosted engine VM.
I'm going to update to 4.1.1.
Normally the workflow was:
- put env in global maintenance - update engine part with engine-setup and such - update other os related packages of hosted engine VM - shutdown hosted engine vm - exit global maintenance
Verify engine VM starts and all is ok form web admin gui. These steps above I have already done and now my engine vm has latest 4.1.1 setup.
Now I want to proceed also with the only existing host part and I have already run:
- shutdown all running VMs - put env in global maintenance - shutdown hosted engine vm
Status is:
[root@ractor ~]# hosted-engine --vm-status
!! Cluster is in GLOBAL MAINTENANCE mode !!
--== Host 1 status ==--
conf_on_shared_storage : True Status up-to-date : True Hostname : ractor.mydomain Host ID : 1 Engine status : {"reason": "bad vm status", "health": "bad", "vm": "down", "detail": "down"} Score : 3400 stopped : False Local maintenance : False crc32 : 017f5635 local_conf_timestamp : 5595857 Host timestamp : 5595835 Extra metadata (valid at timestamp): metadata_parse_version=1 metadata_feature_version=1 timestamp=5595835 (Thu Apr 20 12:06:21 2017) host-id=1 score=3400 vm_conf_refresh_time=5595857 (Thu Apr 20 12:06:44 2017) conf_on_shared_storage=True maintenance=False state=GlobalMaintenance stopped=False
!! Cluster is in GLOBAL MAINTENANCE mode !!
[root@ractor ~]#
Normally I set host into local maintenance now, but I receive this error:
[root@ractor ~]# hosted-engine --set-maintenance --mode=local Unable to enter local maintenance mode: there are no available hosts capable of running the engine VM. [root@ractor ~]#
[root@ractor ~]# ps -ef|grep [k]vm root 887 2 0 Feb14 ? 00:00:00 [kvm-irqfd-clean] [root@ractor ~]#
Is this a bug or changed functionality?
Thanks, Gianluca
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
On Thu, Apr 20, 2017 at 12:12 PM, Gianluca Cecchi <gianluca.cecchi@gmail.com
wrote:
Hello, I have a single host test environment on 4.1.0 with hosted engine VM.
I'm going to update to 4.1.1.
Normally the workflow was:
- put env in global maintenance - update engine part with engine-setup and such - update other os related packages of hosted engine VM - shutdown hosted engine vm - exit global maintenance
Verify engine VM starts and all is ok form web admin gui. These steps above I have already done and now my engine vm has latest 4.1.1 setup.
Now I want to proceed also with the only existing host part and I have already run:
- shutdown all running VMs - put env in global maintenance - shutdown hosted engine vm
Status is:
[root@ractor ~]# hosted-engine --vm-status
!! Cluster is in GLOBAL MAINTENANCE mode !!
--== Host 1 status ==--
conf_on_shared_storage : True Status up-to-date : True Hostname : ractor.mydomain Host ID : 1 Engine status : {"reason": "bad vm status", "health": "bad", "vm": "down", "detail": "down"} Score : 3400 stopped : False Local maintenance : False crc32 : 017f5635 local_conf_timestamp : 5595857 Host timestamp : 5595835 Extra metadata (valid at timestamp): metadata_parse_version=1 metadata_feature_version=1 timestamp=5595835 (Thu Apr 20 12:06:21 2017) host-id=1 score=3400 vm_conf_refresh_time=5595857 (Thu Apr 20 12:06:44 2017) conf_on_shared_storage=True maintenance=False state=GlobalMaintenance stopped=False
!! Cluster is in GLOBAL MAINTENANCE mode !!
[root@ractor ~]#
Normally I set host into local maintenance now, but I receive this error:
[root@ractor ~]# hosted-engine --set-maintenance --mode=local Unable to enter local maintenance mode: there are no available hosts capable of running the engine VM. [root@ractor ~]#
[root@ractor ~]# ps -ef|grep [k]vm root 887 2 0 Feb14 ? 00:00:00 [kvm-irqfd-clean] [root@ractor ~]#
Is this a bug or changed functionality?
Yes in the past it was possible, now it's explicitly prevented since the engine VM couldn't be migrated or restart somewhere else: https://bugzilla.redhat.com/show_bug.cgi?id=1394570
Thanks, Gianluca
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
participants (3)
-
Gianluca Cecchi -
Martin Sivak -
Simone Tiraboschi