4.5.4 Hosted-Engine: change hosted-engine storage
by Diego Ercolani
ovirt-engine-appliance-4.5-20221206133948
Hello,
I have some trouble with my Gluster instance where there is hosted-engine. I would like to copy data from that hosted engine and reverse to another hosted-engine storage (I will try NFS).
I think the main method is to put ovirt in global-management mode, stop the hosted-engine instance, unmount the hosted-engine storage
systemctl stop {vdsmd;supervdsmd;ovirt-ha-broker;ovirt-ha-agent};
hosted-engine --disconnect-storage
umount /rhev/data-center/mnt/glusterSD/localhost:_glen/
and then redeploy the hosted engine with the current backup:
method 1: [doesn't work; it fail the ansible script after the pause]
ovirt-hosted-engine-setup --4 --config-append=/var/lib/ovirt-hosted-engine-setup/answers/answers-20230116184834.conf --restore-from-file=230116-backup.tar.gz --ansible-extra-vars=he_pause_before_engine_setup=true
but it doesn't work
method 2: [I cannot find the glitch]
I saw in every node there is a configuration under /etc/ovirt-hosted-engine/hosted-engine.conf where there are some interessant entry:
vm_disk_id=0a1a501c-fc45-430f-bfd3-076172cec406
vm_disk_vol_id=f65dab86-67f1-46fa-87c0-f9076f479741
storage=localhost:/glen
domainType=glusterfs
metadata_volume_UUID=4e64e155-ee11-4fdc-b9a0-a7cbee5e4181
metadata_image_UUID=20234090-ea40-4614-ae95-8f91b339ba3e
lockspace_volume_UUID=6a975f46-4126-4c2a-b444-6e5a34872cf6
lockspace_image_UUID=893a1fc1-9a1d-44fc-a02f-8fdac19afc18
conf_volume_UUID=206e505f-1bb8-4cc4-abd9-942654c47612
conf_image_UUID=9871a483-8e7b-4f52-bf71-a6a8adc2309b
so, the conf_volume under conf_image dir is a TAR archive:
/rhev/data-center/mnt/glusterSD/localhost:_glen/3577c21e-f757-4405-97d1-0f827c9b4e22/images/9871a483-8e7b-4f52-bf71-a6a8adc2309b/206e505f-1bb8-4cc4-abd9-942654c47612: POSIX tar archive (GNU)
That I think is the common configuration. I copied the structure (from 3577c21e-f757-4405-97d1-0f827c9b4e22 directory) in the new storage (nfs), changed the entries under local configuration and "shared configuration":
from storage=localhost:/glen to storage=<server>:/directory (writable by vdsm user and kvm group)
from domainType=glusterfs to domainType=nfs
and issued:
1. hosted-engine --connect-storage <- work
2. hosted-engine --vm-start <- doesn't work because there is a complain about the not-starting ovirt-ha-agent
What can I do?
Is there any documentation somewhere?
Diego
1 year, 11 months
Bug in the engine-backup script when no attached TTY -- Easy fix
by eshwayri＠gmail.com
The output() function. This line:
printf "%s\n" "${m}"
It will fail if there is no attached TTY, which will set the exit code to 1, which in turn will trigger the cleanup() function notifying the engine that the backup failed. This ironically happens when it should be writing "Done." and exiting after a successful backup. Fix I used was to change it to:
printf "%s\n" "${m}" >> "${LOG}"
You can't assume attached TTY since a lot of people like me want to run this as part of a pre/post script to an automated backup program.
1 year, 11 months
Hosted Engine No Longer Has IP Address - Help!
by Matthew J Black
Hi Guys,
I've gone and shot myself in the foot - and I'm looking for some first-aid.
I've managed to remove the IP Address of the oVirt Self-Hosted Engine and so have lost contact with it (don't ask how - let's just say I f*cked-up). I *think* its still running, I've got it set to DHCP, and I've got access to the Host its running on, so my question(s) is:
- (The preferred method) How can I re-establish (console?) contact - I'm thinking via the Host Server and some kvm-commands, so I can issue a `dhclient` command
- (The most drastic) How can I get it to reboot ie is there a command / command sequence to do this
Any help would be appreciated.
Cheers
Dulux-Oz
1 year, 11 months
update path from 4.4.4 to 4.5.4
by marek
hi,
i have standalone ovirt-engine 4.4.4 (centos8 stream)
few hosts (mix of old centos 8.2, centos 8 stream, rocky linux)
can you confirm if this upgrade path is ok?
first round
- update engine to 4.4.10
- update OS on engine to last centos 8 stream (can i upgrade to some
point in time recommended for 4.4.10?)
- update hosts to ovirt 4.4.10
- update OS on hosts to last centos 8 stream and rocky linux 8.6
special case in first round - centos 8.2
- migrate OS on hosts from centos 8.2 to rocky linux 8.6
- update hosts to ovirt 4.4.10
second round
- update engine to 4.5.4
- update hosts to ovirt 4.5.4
- update OS on hosts - centos 8 stream - is ALREADY on last update from
first round
- update OS on hosts - rocky linux - from 8.6 to 8.7
Marek
1 year, 11 months
Preferred way to give customer access to VMs
by hanisirfan.work＠gmail.com
Hi. Just for an introduction, I'm a junior staff working on a way to deploy a KVM cluster to provision VMs to our customer. Before this, we're using VMware ESXi and connect it to OpenNebula as the console that we give to customers.
We're moving to KVM due to VMware licensing cost. I've successfully deployed an oVirt cluster and currently able to access it remotely via a VPN that I've setup on a virtualized pfSense VM inside the cluster.
My question is, what is the best way to give customer console access to the VMs that we provisioned for them? Surely we doesn't want to give them access to our VPN for security reasons.
I can't seems to find a way to connect OpenNebula with oVirt and I do believe it isn't possible since both are basically virtualization manager, managing the KVM instances.
Thanks for the responses and ideas given ☺️
1 year, 11 months
Failed to import the Hosted Engine Storage Domain
by Frank Wall
Hi,
the following error message is periodically logged to engine.log when
oVirt tries to import the Hosted Engine Storage Domain (NFS):
Validation of action 'AddExistingFileStorageDomain' failed for user
SYSTEM. Reasons:
VAR__TYPE__STORAGE__DOMAIN,VAR__ACTION__ADD,ERROR_CANNOT_ADD_EXISTING_STORAGE_DOMAIN_CONNECTION_DATA_ILLEGAL
I've redeployed my Engine VM while upgrading to oVirt 4.5.
Unfortunately, the Engine VM and Hosted Engine Storage Domain are both
missing from the oVirt GUI and oVirt is unable to import them
apparently. (FWIW, there are several other Storage Domains already
configured that work perfectly fine.)
Is there any way to manually add the Hosted Engine Storage Domain?
I'm also fine with adding the entry manually to the PostgreSQL database
if this would be feasible. (Just point me to the right direction so I
don't mess up my oVirt setup.)
Below are some log messages and errors I've collected from vdsm.log and
engine.log.
vdsm.log
2023-01-06 21:15:10,803+0100 INFO (jsonrpc/0) [vdsm.api] FINISH
getStorageDomainInfo return={'info': {'uuid':
'49d7aab9-1e45-4ef0-9657-1247983f129e', 'type': 'NFS', 'class': 'Data',
'name': 'hosted_storage', 'role': 'Master', 'pool':
['c87a1ad0-8dcd-11ed-b48a-00163e352102'], 'version': '5', 'block_size':
512, 'alignment': 1048576, 'remotePath':
'10.1.2.3:/mnt/vol2/vm/engine'}} from=::ffff:192.168.0.30,43146,
flow_id=4ba364c1, task_id=43840eb6-8b9e-4637-bd0f-261a899d374e (api:37)
engine.log
2023-01-06 21:16:28,042+01 INFO
[org.ovirt.engine.core.vdsbroker.vdsbroker.HSMGetStorageDomainInfoVDSCommand]
(EE-ManagedThreadFactory-engine-Thread-1342) [] FINISH,
HSMGetStorageDomainInfoVDSCommand, return:
<StorageDomainStatic:{name='hosted_storage',
id='49d7aab9-1e45-4ef0-9657-1247983f129e'},
c87a1ad0-8dcd-11ed-b48a-00163e352102>, log id: 5344c976
2023-01-06 21:16:28,046+01 INFO
[org.ovirt.engine.core.bll.storage.domain.ImportHostedEngineStorageDomainCommand]
(EE-ManagedThreadFactory-engine-Thread-1342) [1728fc85] Running command:
ImportHostedEngineStorageDomainCommand internal: true.
2023-01-06 21:16:28,086+01 INFO
[org.ovirt.engine.core.vdsbroker.vdsbroker.HSMGetStorageDomainInfoVDSCommand]
(EE-ManagedThreadFactory-engine-Thread-1342) [1d6e287c] START,
HSMGetStorageDomainInfoVDSCommand(HostName = hv05.example.com,
HSMGetStorageDomainInfoVDSCommandParameters:{hostId='4fa6fe39-9254-4264-8a57-4b0a29c645e7',
storageDomainId='49d7aab9-1e45-4ef0-9657-1247983f129e'}), log id:
759c4a73
2023-01-06 21:16:28,095+01 INFO
[org.ovirt.engine.core.vdsbroker.vdsbroker.HSMGetStorageDomainInfoVDSCommand]
(EE-ManagedThreadFactory-engine-Thread-1342) [1d6e287c] FINISH,
HSMGetStorageDomainInfoVDSCommand, return:
<StorageDomainStatic:{name='hosted_storage',
id='49d7aab9-1e45-4ef0-9657-1247983f129e'},
c87a1ad0-8dcd-11ed-b48a-00163e352102>, log id: 759c4a73
2023-01-06 21:16:28,095+01 WARN
[org.ovirt.engine.core.bll.storage.domain.AddExistingFileStorageDomainCommand]
(EE-ManagedThreadFactory-engine-Thread-1342) [1d6e287c] Validation of
action 'AddExistingFileStorageDomain' failed for user SYSTEM. Reasons:
VAR__TYPE__STORAGE__DOMAIN,VAR__ACTION__ADD,ERROR_CANNOT_ADD_EXISTING_STORAGE_DOMAIN_CONNECTION_DATA_ILLEGAL
2023-01-06 21:16:28,096+01 INFO
[org.ovirt.engine.core.bll.storage.domain.AddExistingFileStorageDomainCommand]
(EE-ManagedThreadFactory-engine-Thread-1342) [1d6e287c] Lock freed to
object
'EngineLock:{exclusiveLocks='[49d7aab9-1e45-4ef0-9657-1247983f129e=STORAGE]',
sharedLocks=''}'
2023-01-06 21:16:28,097+01 INFO
[org.ovirt.engine.core.bll.CommandCompensator]
(EE-ManagedThreadFactory-engine-Thread-1342) [1d6e287c] Command
[id=b0a331fe-5ad5-40b0-b7ce-2d01d0276e4b]: Compensating
DELETED_OR_UPDATED_ENTITY of
org.ovirt.engine.core.common.businessentities.StorageServerConnections;
snapshot: id=c346dafe-09db-4058-ae1c-9b609911a1af.
2023-01-06 21:16:28,100+01 ERROR
[org.ovirt.engine.core.bll.storage.domain.ImportHostedEngineStorageDomainCommand]
(EE-ManagedThreadFactory-engine-Thread-1342) [1d6e287c] Transaction
rolled-back for command
'org.ovirt.engine.core.bll.storage.domain.ImportHostedEngineStorageDomainCommand'.
The (new) Engine VM is running on CentOS Stream 8 (using the Engine
Appliance from the rpm) and all oVirt nodes/hosts are running on CentOS
Stream 9.
Ciao
- Frank
1 year, 11 months
Re: aggregate multiple hosts
by Nathanaël Blanchet
Le 13/01/2023 à 10:57, Csókási István Lehel a écrit :
>
> I want to increase the performance of a virtual machine by connecting
> 6 nodes. Windows 10 runs on the virtual machine, the nodes are HP
> Compaq 7900 PCs with Intel E8400 3 GHz Dual core processor, Q45
> chipset, 4Gb RAM
>
> Thank you
>
If you want to aggregate hardware capabilities to an unique VM, i.e.
vcpu from multiple hosts or RAM aggragation, I don't think it is
possible, but I may be wrong... some other members of the community may
confirm.
>
>
> 2023. 01. 13. 10:18 keltezéssel, Nathanaël Blanchet írta:
>> You should precise your intention to do so.
>> Thank you
>>
>> Le 13 janv. 2023 08:52, csokasi.lehel(a)gmail.com a écrit :
>> I would like to ask if it is possible to aggregate multiple hosts to
>> a single virtual machine.
>> _______________________________________________
>> Users mailing list -- users(a)ovirt.org
>> To unsubscribe send an email to users-leave(a)ovirt.org
>> Privacy Statement: https://www.ovirt.org/privacy-policy.html
>> oVirt Code of Conduct:
>> https://www.ovirt.org/community/about/community-guidelines/
>> List Archives:
>> https://lists.ovirt.org/archives/list/users@ovirt.org/message/665K3Y7QSIN...
1 year, 11 months
interface http stuck
by remy.ressegaire＠unige.ch
Hello,
When browsing our ovirtadmin web interface it never loads the resource.
We are using a third party certificate authority (following this procedure https://myhomelab.gr/linux/2020/01/20/replacing_ovirt_ssl.html), but not sure this leads to the error we have.
In the http logs I see :
ssl_error_log :
[proxy_ajp:error] [pid 23754] [client 10.25.64.40:57838] AH00992: ajp_read_header: ajp_ilink_receive failed
[proxy_ajp:error] [pid 23754] (120006)APR does not understand this error code: [client 10.25.64.40:57838] AH00878: read response failed from 127.0.0.1:8702 (127.0.0.1)
error.log :
[proxy:error] [pid 20752] (111)Connection refused: AH00957: AJP: attempt to connect to 127.0.0.1:8702 (127.0.0.1) failed
[proxy:error] [pid 20752] AH00959: ap_proxy_connect_backend disabling worker for (127.0.0.1) for 5s
[proxy_ajp:error] [pid 20752] [client 10.25.64.44:50000] AH00896: failed to make connection to backend: 127.0.0.1
Any help will be welcomed.
Thanks,
Remo
1 year, 11 months
