So, let's summarize:
- Cannot migrate the HE due to "CPU policy".
- HE's CPU is westmere - just like hosts
- You have enough resources on the second HE host (both CPU + MEMORY)
What is the Cluster's CPU type (you can check in UI) ?
Maybe you should enable debugging on various locations to identi
Yes. I can. The host which does not host the HE could be reinstalled
sucessfully in web UI. After this is done nothing has changed.
在 2020-09-22 03:08:18,"Strahil Nikolov" 写道:
>Can you put 1 host in maintenance and use the "Installation" -> "Reinstall"
>and enable the HE deployment
Can you put 1 host in maintenance and use the "Installation" -> "Reinstall" and
enable the HE deployment from one of the tabs ?
Best Regards,
Strahil Nikolov
В понеделник, 21 септември 2020 г., 06:38:06 Гринуич+3, ddqlo
написа:
so strange! After I set global maintenance, powered off
That's quite strange.
Any errors/clues in the Engine's logs ?
Best Regards,
Strahil Nikolov
В понеделник, 21 септември 2020 г., 05:58:35 Гринуич+3, ddqlo
написа:
so strange! After I set global maintenance, powered off and started H The cpu
of HE became 'Westmere'(did not change anyt
so strange! After I set global maintenance, powered off and started H The cpu
of HE became 'Westmere'(did not change anything). But HE still could not be
migrated.
HE xml:
Westmere
host capabilities:
Westmere
cluster cpu type (UI):
Hm... interesting.
The VM is using 'Haswell-noTSX' while the host is 'Westmere'.
In my case I got no difference:
[root@ovirt1 ~]# virsh dumpxml HostedEngine | grep Opteron
Opteron_G5
[root@ovirt1 ~]# virsh capabilities | grep Opteron
Opteron_G5
Did you update the cluster holding the H
HE:
HostedEngine
b4e805ff-556d-42bd-a6df-02f5902fd01c
http://ovirt.org/vm/tune/1.0";
xmlns:ovirt-vm="http://ovirt.org/vm/1.0";>
http://ovirt.org/vm/1.0";>
4.3
False
false
1024
1024
auto_resume
1600307555.19
external
4
It would be easier if you posted the whole xml.
What about the sections (in HE xml) starting with:
feature policy=
Also the hosts have a section which contains:
написа:
HostedEngine:
..
Haswell-noTSX
..
both of the hosts:
..
Westmere
..
others vms which can be migrated:
It would be easier if you posted the whole xml.
What about the sections (in HE xml) starting with:
feature policy=
Also the hosts have a section which contains:
В четвъртък, 17 септември 2020 г., 05:54:12 Гринуич+3, ddqlo
написа:
HostedEngine:
..
Haswell-noTSX
..
both of
HostedEngine:
..
Haswell-noTSX
..
both of the hosts:
..
Westmere
..
others vms which can be migrated:
..
Haswell-noTSX
..
在 2020-09-17 03:03:24,"Strahil Nikolov" 写道:
>Can you verify the HostedEngine's CPU ?
>
>1. ssh to the host hosting the HE
>2. alias virsh='virs
Can you verify the HostedEngine's CPU ?
1. ssh to the host hosting the HE
2. alias virsh='virsh -c
qemu:///system?authfile=/etc/ovirt-hosted-engine/virsh_auth.conf'
3. virsh dumpxml HostedEngine
Then set the alias for virsh on all Hosts and 'virsh capabilites' should show
the Hosts' .
Best R
My gateway was not pingable. I have fixed this problem and now both nodes have
a score(3400).
Yet, hosted engine could not be migrated. Same log in engine.log:
host filtered out by 'VAR__FILTERTYPE__INTERNAL' filter 'CPU'
在 2020-09-16 02:11:09,"Strahil Nikolov" 写道:
>Both nodes have a lowe
Both nodes have a lower than the usual score (should be 3400 ).
Based on the score you are probably suffering from gateway-score-penalty [1][2].
Check if your gateway is pingable.
Best Regards,
Strahil Nikolov
1 - https://www.ovirt.org/images/Hosted-Engine-4.3-deep-dive.pdf(page 8)
2 - /etc/ovirt
I have tried. It does not work.
My host network:
Will this help?
在 2020-09-09 22:52:25,"Strahil Nikolov" 写道:
>I think that you can try to set one of the HE hosts into maintenance and then
>use UI to 'reinstall'. Don't forget to mark the host as a HE host also (some
>dropdown in the U
--== Host node28 (id: 1) status ==--
conf_on_shared_storage : True
Status up-to-date : True
Hostname : node28
Host ID: 1
Engine status : {"reason": "vm not running on this host",
"health
diffidences:
a15b30fd-2de2-4bea-922d-d0de2ee3b76a
..
..
32904772
8226193
在 2020-09-09 01:35:59,"Strahil Nikolov" 写道:
>You can use the following:
>
>vim ~/.bashrc
>
>alias virsh='virsh -c
>qemu:///system?authfile=/etc/
I think that you can try to set one of the HE hosts into maintenance and then
use UI to 'reinstall'. Don't forget to mark the host as a HE host also (some
dropdown in the UI wizard).
Best Regards,
Strahil Nikolov
В вторник, 8 септември 2020 г., 10:24:00 Гринуич+3, Yedidyah Bar David
напи
You can use the following:
vim ~/.bashrc
alias virsh='virsh -c
qemu:///system?authfile=/etc/ovirt-hosted-engine/virsh_auth.conf'
source ~/.bashrc
#Show host capabilities
virsh capabilities
Now repeat on the other nodes. Compare the CPU from the 3 outputs.
Best Regards,
Strahil Nikolov
What is the output of 'hosted-engine --vm-status' on the node where the
HostedEngine is running ?
Best Regards,
Strahil Nikolov
В понеделник, 7 септември 2020 г., 03:53:13 Гринуич+3, ddqlo
написа:
I could not find any logs because the migration button is disabled in the web
UI. It
On Tue, Sep 8, 2020 at 4:33 AM ddqlo wrote:
> my hosts cpu:Intel Haswell-noTSX Family
>
> cluster cpu:Intel Haswell-noTSX Family
> HostedEngine vm cpu:Intel Haswell-noTSX Family
>
> When I tried to put the host in maintenance in web UI, I got an error:
>
>
Adding Arik. Arik - any idea what else t
my hosts cpu:Intel Haswell-noTSX Family
cluster cpu:Intel Haswell-noTSX Family
HostedEngine vm cpu:Intel Haswell-noTSX Family
When I tried to put the host in maintenance in web UI, I got an error:
When I typed the command, I got this:
[root@node22 ~]# hosted-engine --set-maintenance --mode
On Mon, Sep 7, 2020 at 4:13 AM ddqlo wrote:
>
> I have found some engine logs:
>
> 2020-09-07 09:00:45,428+08 INFO
> [org.ovirt.engine.core.bll.scheduling.SchedulingManager] (default task-6)
> [29259482-1515-4c10-8458-59354a0953ac] Candidate host 'node22'
> ('585b374b-4c82-4f5c-aad7-196d9f5d56
I have found some engine logs:
2020-09-07 09:00:45,428+08 INFO
[org.ovirt.engine.core.bll.scheduling.SchedulingManager] (default task-6)
[29259482-1515-4c10-8458-59354a0953ac] Candidate host 'node22'
('585b374b-4c82-4f5c-aad7-196d9f5d5625') was filtered out by
'VAR__FILTERTYPE__INTERNAL' fil
I could not find any logs because the migration button is disabled in the web
UI. It seems that the engine migration operation is prevented at first. Any
other ideas? Thanks!
在 2020-09-01 00:06:19,"Strahil Nikolov" 写道:
>I'm running oVirt 4.3.10 and I can migrate my Engine from n
I'm running oVirt 4.3.10 and I can migrate my Engine from node to node.
I had one similar issue , but powering off and on the HE has fixed it.
You have to check the vdsm log on the source and on destination in order to
figure out what is going on.
Also you might consider checking the libvirt logs
Thanks! The scores of all nodes are not '0'. I find that someone has already
asked a question like this. It seems that this feature has been disabled in
4.3. I am not sure if it is enabled in 4.4.
在 2020-08-29 02:27:03,"Strahil Nikolov" :
>Have you checked under a shell the output of 'hoste
Have you checked under a shell the output of 'hosted-engine --vm-status' .
Check the Score of the hosts. Maybe there is a node with score of '0' ?
Best Regards,
Strahil Nikolov
В вторник, 25 август 2020 г., 13:46:18 Гринуич+3, 董青龙 написа:
Hi all,
I have an ovirt4.3.10.4 envir
I tend to go through and check everything possible. I check all the
features of a fresh system before I start to add new VM's. So here is
the trouble I am having then.
I have a 3 node cluster. Each node is running Ovirt-Node iso ver 4.3.0
upgraded to version 4.3.1. Hosted engine matches version. T
It seems my first post didn't go through correctly, probably my fault. Here is
the information, just incase it was a complete fail.
I tend to go through and check everything possible. I check all the
features of a fresh system before I start to add new VM's. So here is
the trouble I am having the
Also, it seems that whichever host I am attempting to migrate to volume
data, engine and vmstore seem to go down
On Mon, 2019-03-11 at 14:10 +0100, Martin Sivak wrote:
> Hi,
>
> as far as I know you can manually migrate hosted engine from the
> webadmin UI by clicking at the migrate button.
> T
Hi,
as far as I know you can manually migrate hosted engine from the
webadmin UI by clicking at the migrate button.
The question is, why would you want to?
Best regards
Martin Sivak
On Mon, Mar 11, 2019 at 2:01 PM wrote:
>
> Am I reading these right in that manual migration is not possible?
>
31 matches
Mail list logo