Re: [ovirt-users] iSCSI Multipathing -> host inactive

Hi, sorry for the delay, I reinstalled everything, configured the networks, attached the iSCSI storage with 2 interfaces and finally created the iSCSI-bond: [root@ovh01 ~]# route > Kernel IP Routentabelle > Ziel Router Genmask Flags Metric Ref Use > Iface > default hp5406-1-srv.mo 0.0.0.0 UG 0 0 0 > ovirtmgmt > 10.0.24.0 0.0.0.0 255.255.255.0 U 0 0 0 > ovirtmgmt > 10.0.131.0 0.0.0.0 255.255.255.0 U 0 0 0 > enp9s0f0 > 10.0.132.0 0.0.0.0 255.255.255.0 U 0 0 0 > enp9s0f1 > link-local 0.0.0.0 255.255.0.0 U 1005 0 0 > enp9s0f0 > link-local 0.0.0.0 255.255.0.0 U 1006 0 0 > enp9s0f1 > link-local 0.0.0.0 255.255.0.0 U 1008 0 0 > ovirtmgmt > link-local 0.0.0.0 255.255.0.0 U 1015 0 0 > bond0 > link-local 0.0.0.0 255.255.0.0 U 1017 0 0 > ADMIN > link-local 0.0.0.0 255.255.0.0 U 1021 0 0 > SRV > and: [root@ovh01 ~]# ip a > 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN > link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 > inet 127.0.0.1/8 scope host lo > valid_lft forever preferred_lft forever > inet6 ::1/128 scope host > valid_lft forever preferred_lft forever > 2: enp13s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast > master ovirtmgmt state UP qlen 1000 > link/ether e0:3f:49:6d:68:c4 brd ff:ff:ff:ff:ff:ff > 3: enp8s0f0: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 1500 qdisc mq > master bond0 state UP qlen 1000 > link/ether 90:e2:ba:11:21:d0 brd ff:ff:ff:ff:ff:ff > 4: enp8s0f1: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 1500 qdisc mq > master bond0 state UP qlen 1000 > link/ether 90:e2:ba:11:21:d0 brd ff:ff:ff:ff:ff:ff > 5: enp9s0f0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state > UP qlen 1000 > link/ether 90:e2:ba:11:21:d4 brd ff:ff:ff:ff:ff:ff > inet 10.0.131.181/24 brd 10.0.131.255 scope global enp9s0f0 > valid_lft forever preferred_lft forever > 6: enp9s0f1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state > UP qlen 1000 > link/ether 90:e2:ba:11:21:d5 brd ff:ff:ff:ff:ff:ff > inet 10.0.132.181/24 brd 10.0.132.255 scope global enp9s0f1 > valid_lft forever preferred_lft forever > 7: ;vdsmdummy;: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN > link/ether 26:b2:4e:5e:f0:60 brd ff:ff:ff:ff:ff:ff > 8: ovirtmgmt: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue > state UP > link/ether e0:3f:49:6d:68:c4 brd ff:ff:ff:ff:ff:ff > inet 10.0.24.181/24 brd 10.0.24.255 scope global ovirtmgmt > valid_lft forever preferred_lft forever > 14: vnet0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast > master ovirtmgmt state UNKNOWN qlen 500 > link/ether fe:16:3e:79:25:86 brd ff:ff:ff:ff:ff:ff > inet6 fe80::fc16:3eff:fe79:2586/64 scope link > valid_lft forever preferred_lft forever > 15: bond0: <BROADCAST,MULTICAST,MASTER,UP,LOWER_UP> mtu 1500 qdisc > noqueue state UP > link/ether 90:e2:ba:11:21:d0 brd ff:ff:ff:ff:ff:ff > 16: bond0.32@bond0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc > noqueue master ADMIN state UP > link/ether 90:e2:ba:11:21:d0 brd ff:ff:ff:ff:ff:ff > 17: ADMIN: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue > state UP > link/ether 90:e2:ba:11:21:d0 brd ff:ff:ff:ff:ff:ff > 20: bond0.24@bond0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc > noqueue master SRV state UP > link/ether 90:e2:ba:11:21:d0 brd ff:ff:ff:ff:ff:ff > 21: SRV: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state > UP > link/ether 90:e2:ba:11:21:d0 brd ff:ff:ff:ff:ff:ff > The host keeps toggling all storage domains on and off as soon as there is an iSCSI bond configured. Thank you for your patience. cu, Uwe Am 18.08.2016 um 11:10 schrieb Elad Ben Aharon: > I don't think it's necessary. > Please provide the host's routing table and interfaces list ('ip a' or > ifconfing) while it's configured with the bond. > > Thanks > > On Tue, Aug 16, 2016 at 4:39 PM, Uwe Laverenz <uwe(a)laverenz.de > <mailto:uwe@laverenz.de>> wrote: > > Hi Elad, > > Am 16.08.2016 um 10:52 schrieb Elad Ben Aharon: > > Please be sure that ovirtmgmt is not part of the iSCSI bond. > > > Yes, I made sure it is not part of the bond. > > It does seem to have a conflict between default and enp9s0f0/ > enp9s0f1. > Try to put the host in maintenance and then delete the iscsi > nodes using > 'iscsiadm -m node -o delete'. Then activate the host. > > > I tried that, I managed to get the iSCSI interface clean, no > "default" anymore. But that didn't solve the problem of the host > becoming "inactive". Not even the NFS domains would come up. > > As soon as I remove the iSCSI-bond, the host becomes responsive > again and I can activate all storage domains. Removing the bond also > brings the duplicated "Iface Name" back (but this time causes no > problems). > > ... > > I wonder if there is a basic misunderstanding on my side: wouldn't > it be necessary that all targets are reachable from all interfaces > that are configured into the bond to make it work? > > But this would either mean two interfaces in the same network or > routing between the iSCSI networks. > > Thanks, > > Uwe > _______________________________________________ > Users mailing list > Users(a)ovirt.org <mailto:Users@ovirt.org> > http://lists.ovirt.org/mailman/listinfo/users > <http://lists.ovirt.org/mailman/listinfo/users> > > > _______________________________________________ Users mailing list Users(a)ovirt.org http://lists.ovirt.org/mailman/listinfo/users