Bug 1655375

It's a code problem, but in the meantime, is there anything unusual about the nics? A field null that usually isn't?



2018-12-03 11:37:46,114+10 ERROR [org.ovirt.engine.ui.frontend.server.gwt.OvirtRemoteLoggingService] (default task-20) [] Uncaught exception: com.google.gwt.core.client.JavaScriptException: (TypeError) : Cannot read property 'a' of null
	at org.ovirt.engine.ui.webadmin.widget.host.HostNetworkInterfaceListViewItem.$createBodyPanel(HostNetworkInterfaceListViewItem.java:124)
	at org.ovirt.engine.ui.webadmin.widget.host.HostNetworkInterfaceListViewItem.createBodyPanel(HostNetworkInterfaceListViewItem.java:124)
	at org.ovirt.engine.ui.common.widget.listgroup.PatternflyListViewItem.PatternflyListViewItem(PatternflyListViewItem.java:77)
	at org.ovirt.engine.ui.webadmin.widget.host.HostNetworkInterfaceListViewItem.HostNetworkInterfaceListViewItem(HostNetworkInterfaceListViewItem.java:76)
	at Unknown.new sDy(webadmin-66.js)
	at org.ovirt.engine.ui.webadmin.section.main.view.tab.host.SubTabHostInterfaceView.$createListViewItem(SubTabHostInterfaceView.java:70)
	at org.ovirt.engine.ui.webadmin.section.main.view.tab.host.SubTabHostInterfaceView.createListViewItem(SubTabHostInterfaceView.java:70)
	at org.ovirt.engine.ui.common.widget.listgroup.PatternflyListView.$updateInfoPanel(PatternflyListView.java:137)
	at org.ovirt.engine.ui.common.widget.listgroup.PatternflyListView.$lambda$1(PatternflyListView.java:63)
	at org.ovirt.engine.ui.common.widget.listgroup.PatternflyListView$lambda$1$Type.eventRaised(PatternflyListView.java:63)
	at org.ovirt.engine.ui.uicompat.Event.$raise(Event.java:99)
	at org.ovirt.engine.ui.uicommonweb.models.hosts.HostInterfaceListModel.$setItems(HostInterfaceListModel.java:94)
	at org.ovirt.engine.ui.uicommonweb.models.hosts.HostInterfaceListModel.$updateItems(HostInterfaceListModel.java:188)
	at org.ovirt.engine.ui.uicommonweb.models.hosts.HostInterfaceListModel.$lambda$1(HostInterfaceListModel.java:182)
	at org.ovirt.engine.ui.uicommonweb.models.hosts.HostInterfaceListModel$lambda$1$Type.onSuccess(HostInterfaceListModel.java:182)
	at org.ovirt.engine.ui.frontend.Frontend$1.$onSuccess(Frontend.java:227) [frontend.jar:]

@Ales, this may be a bug or limitation in PatternflyListView. The results of this search are telling:

https://www.google.com/search?q=inurl%3Abugzilla.redhat.com+PatternflyListView

OK, I have a clue on why this came up:

1) We know 4.2 admin portal bombs out if speed is null
2) 4.1 vdsm does not report the speed for bond/vlan devices in getCaps

Apparently it was added in 4.2 vdsm via this patch?    
net: Report bond (virtual) speed through net caps

I can reproduce the problem by using:
* vdsm-4.19.31
* ovirt-engine-4.2.7
* configure a bond using 2 NICs.

No matter how many Refresh Capabilities I do, the speed for bond devices is always NULL and the UI throws the exact same exception.

vdsm-4.20.44:

    "bondings": {
        "bond0": {
            "ipv6autoconf": true, 
            "addr": "192.168.101.3", 
            "speed": 0,                   <---------------
            "dhcpv6": false, 
            "ipv6addrs": [], 
            "netmask": "255.255.255.0", 
            "active_slave": "eth1", 
            "mtu": "1500", 
            "dhcpv4": true, 
            "switch": "legacy", 
            "ipv4defaultroute": false, 
            "ipv4addrs": [
                "192.168.101.3/24"
            ], 
            "hwaddr": "52:54:00:16:c1:03", 
            "slaves": [
                "eth2", 
                "eth1"
            ], 
            "ipv6gateway": "::", 
            "gateway": "", 
            "opts": {
                "miimon": "100", 
                "mode": "1"
            }
        }
    }, 


vdsm-4.19.31:

        "bond0": {
            "ipv6autoconf": true, 
            "addr": "192.168.101.3", 
            "ipv4defaultroute": false,    <----- no speed here
            "ipv6addrs": [], 
            "switch": "legacy", 
            "active_slave": "eth1", 
            "mtu": "1500", 
            "dhcpv4": true, 
            "netmask": "255.255.255.0", 
            "dhcpv6": false, 
            "ipv4addrs": [
                "192.168.101.3/24"
            ], 
            "hwaddr": "52:54:00:16:c1:03", 
            "slaves": [
                "eth1", 
                "eth2"
            ], 
            "ipv6gateway": "::", 
            "gateway": "", 
            "opts": {
                "miimon": "100", 
                "mode": "1"
            }
        }
    },

This affects anybody upgrading Engine to 4.2 before the hosts, asking for exception.

Yes, I think that this bug needs to be cloned to 4.2.z and backported.

I hope that Dominik can do both tomorrow if Martin acks it.

Verified on - rhvm-4.3.2-0.1.el7.noarch with vdsm-4.30.10-1.el7ev.x86_64

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHEA-2019:1085