[ovirt-users] Network statistics shown in ovirt doesn't look correct.
Dan Kenigsberg
danken at redhat.com
Tue Sep 8 12:07:04 UTC 2015
On Tue, Sep 08, 2015 at 04:22:27PM +0530, Ramesh Nachimuthu wrote:
>
>
> On 09/07/2015 05:47 PM, Dan Kenigsberg wrote:
> >On Mon, Sep 07, 2015 at 04:51:34PM +0530, Ramesh Nachimuthu wrote:
> >>Hi,
> >>
> >> I have a strange issue with the network traffic shown in oVirt webadmin
> >>portal. I have 2 10 Gb network card and I have created bond out of them. I
> >>have used *'iperf*' command to generate10 Gbps traffic and I can see that
> >>getting reflected in my nagios monitoring. Also iperf command confirms that
> >>it transferred the data at the rate of 9.38 Gbits/sec. But oVirt Ui shows
> >>only 1500 Mbps. I am not sure how oVirt shows very low traffic. Anyone
> >>experiencing the similar problem?
> >>
> >>VDSM Version: vdsm-4.16.20-1.3.el6rhs.x86_64
> >>Ovirt release : both 3.5 and 3.6
> >>
> >>I have attached the relevant screen shots here.
> >Can you provide the output of
> >
> > vdsClient -s 0 getVdsStats
> >
> >when iperf is running?
>
> Here is the output of vdsClient
> http://ur1.ca/nporl
'bond0': {'name': 'bond0',
'rxDropped': '0',
'rxErrors': '0',
'rxRate': '7.1',
'speed': '20000',
'state': 'up',
'txDropped': '0',
'txErrors': '0',
'txRate': '0.3'},
'ovirtmgmt': {'name': 'ovirtmgmt',
'rxDropped': '0',
'rxErrors': '0',
'rxRate': '100.0',
'speed': '1000',
'state': 'up',
'txDropped': '0',
'txErrors': '0',
'txRate': '5.5'}}
This may very well be a case of
Bug 1120778 - align reported speeds for physical bond and bridge
but I'm not sure when we are going to fix it, since 3.6 vdsm reports absolute
rx/tx values, which lets Engine provide more sensible data to the user.
>
> >
> >Could you see if a similar issue happens with a new 3.6 vdsm?
>
> I have to re-install the machine with 3.6. I will update u after doing the
> same.
More information about the Users
mailing list