I have 2 virtual machines, one with groupwise 2014 r2 and mobility on the other. They were both working great until a few days ago… Both are showing excessive dropped packets. I added a new nic card and changed the bridge to the nic card, but the dropped packets continue, groupwise is very unstable when sending emails, sometimes it sends, sometimes the emails are delayed… I checked everything and have concluded that the software for the bridged interface on the xen is the issue, but I don´t see that indicate a fix or an update for the bridge… I have been trying to resolve this with Novell, but they don´t have a clue… Any help is appreciated…
Hi crobinson,
Both are showing excessive dropped packets.
where exactly do you see the dropped packets - on the virtual interface, inside the VMs? How do you measure that value, do you have any more details on this?
That the software should act up suddenly, does not seem very likely to me. I’d rather expect other circumstances to apply. You have been exchanging the network card already, are there any other (external) changes that took place around the first incarnation of this problem? Do you have network traces to compare what packets are being dropped?
There may be a multitude of causes for this, but from your report I take it that the base server itself is not reporting dropped packets.
Two shots into the dark:
-
Might it be the MTU sizes differ across the various interfaces (Dom0 LAN interface, bridge interface, DomU interface(s))?
-
Was “huge packet support” introduced to your network prior to seeing the dropped packets? IIRC that bridge code does not support this.
Regards,
Jens
I notice that the bridge is the problem, because from the host of the vm, I can navigate quickly and the packet loss is small. But when I try to navigate from the virtual machines, it is impossible…
And how about MTU and packet size changes?
Regards,
J
MTU 1500 all around…
I think this has to do with the drivers of the nic cards and the SuSE 11 sp2 with the virtualization… I am thinking of updating the SuSE SLES Server to sp3 to see if this resolves the issue… I read a lot of issues with realtek nic drivers and SuSE 11 sp2 and the kernel…
Sorry for the late reply - if you’re considering updating anyhow, why not go all the way to a currently supported level, like SLES11SP4 (or SLES12SP1, although much has changed, which might make this a too big step for now)?
Regards,
Jens