11:49 p.m.
Hi,
I'm trying to set up ovirt using the hosted-engine --deploy command on CentOS7, but am
encountering an error. I am running a slightly unusual network configuration. I have two
fairly basic non stacked gigabit switches with port channels connecting the two switches
together. I have a lacp bond from the host consisting of 4 ports to each switch (bond1 and
bond2). I have then created an active-backup bond (bond0) using the two lacp bonds as
slaves in the hope to create ha at the switch layer using my basic switches. There is then
a VLAN (101) on bond0.
This network configuration runs fine on the host, however, when run, after a short while,
the hosted-engine --deploy command outputs the following error:
...
[ INFO ] TASK [Force host-deploy in offline mode]
[ INFO ] ok: [localhost]
[ INFO ] TASK [Add host]
[ INFO ] changed: [localhost]
[ INFO ] TASK [Wait for the host to be up]
[ INFO ] ok: [localhost]
[ INFO ] TASK [Check host status]
[ ERROR ] fatal: [localhost]: FAILED! => {"changed": false, "msg":
"The host has been set in non_operational status, please check engine logs, fix
accordingly and re-deploy.\n"}
...
Looking in /var/log/ovirt-engine/engine.log on the machine created, I can see the
following errors logged:
...
2018-10-04 21:51:30,116+01 INFO
[org.ovirt.engine.core.vdsbroker.vdsbroker.HostSetupNetworksVDSCommand]
(EE-ManagedThreadFactory-engine-Thread-1) [59fb360a] START,
HostSetupNetworksVDSCommand(HostName = ov1.test.local,
HostSetupNetworksVdsCommandParameters:{hostId='7440c9b9-e530-4341-a317-d3a9041dc777',
vds='Host[ov1.test.local,7440c9b9-e530-4341-a317-d3a9041dc777]',
rollbackOnFailure='true', connectivityTimeout='120',
networks='[HostNetwork:{defaultRoute='true', bonding='true',
networkName='ovirtmgmt', vdsmName='ovirtmgmt', nicName='bond0',
vlan='101', vmNetwork='true', stp='false',
properties='null', ipv4BootProtocol='STATIC_IP',
ipv4Address='192.168.1.11', ipv4Netmask='255.255.255.0',
ipv4Gateway='192.168.1.1', ipv6BootProtocol='AUTOCONF',
ipv6Address='null', ipv6Prefix='null', ipv6Gateway='null',
nameServers='null'}]', removedNetworks='[]', bonds='[]',
removedBonds='[]', clusterSwitchType='LEGACY'}), log id: 4f0c7eaa
2018-10-04 21:51:30,121+01 INFO
[org.ovirt.engine.core.vdsbroker.vdsbroker.HostSetupNetworksVDSCommand]
(EE-ManagedThreadFactory-engine-Thread-1) [59fb360a] FINISH, HostSetupNetworksVDSCommand,
log id: 4f0c7eaa
2018-10-04 21:51:30,645+01 ERROR
[org.ovirt.engine.core.vdsbroker.vdsbroker.HostSetupNetworksVDSCommand]
(EE-ManagedThreadFactory-engine-Thread-1) [59fb360a] Failed in
'HostSetupNetworksVDS' method
2018-10-04 21:51:30,687+01 ERROR
[org.ovirt.engine.core.dal.dbbroker.auditloghandling.AuditLogDirector]
(EE-ManagedThreadFactory-engine-Thread-1) [59fb360a] EVENT_ID:
VDS_BROKER_COMMAND_FAILURE(10,802), VDSM ov1.test.local command HostSetupNetworksVDS
failed: Unknown nics in: ['bond1', 'bond2']
2018-10-04 21:51:30,688+01 ERROR
[org.ovirt.engine.core.vdsbroker.vdsbroker.HostSetupNetworksVDSCommand]
(EE-ManagedThreadFactory-engine-Thread-1) [59fb360a] Error: VDSGenericException:
VDSErrorException: Failed to HostSetupNetworksVDS, error = Unknown nics in:
['bond1', 'bond2'], code = 23
...
It looks like when HostSetupNetworksVDS is run, it is checking that the slave interfaces
to the bonds are physical network devices and being as the slaves of bond0 are bond1 and
bond2, rather than physical devices, it then throws the error Unknown nics in:
['bond1', 'bond2'].
Is there anything I can do or any configuration that I can put anywhere to make it work
with this "stacked bond" configuration or does ovirt just not work when bonds
are set up like this?
Thanks in advance,
Ben