[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-users] Strange failures of Xen 4.3.1, PVHVM storage VM, iSCSI and Windows+GPLPV VM combination


  • To: xen-users@xxxxxxxxxxxxx
  • From: Kuba <kuba.0000@xxxxx>
  • Date: Thu, 06 Feb 2014 11:14:50 +0100
  • Delivery-date: Thu, 06 Feb 2014 10:15:19 +0000
  • List-id: Xen user discussion <xen-users.lists.xen.org>

W dniu 2014-02-06 00:07, James Harper pisze:


Any ideas and suggestions on what could be the cause and what can I do
will be much appreciated.


Can you try turning on iscsi data checksuming? If data corruption in the 
network stack is the cause (seems likely to me) then this should pick it up.

If your iscsi initiator and/or target don't support data digest, try turning 
off all the offloads and scatter gather on the network interfaces and see if 
you can reduce the problem to one of the offloads (large send would be my best 
guess).

James
_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxx
http://lists.xen.org/xen-users



I haven't noticed any change after turning off all offloads etc. reported by ethtool. Sorry, I forgot to mention that earlier.

I couldn't get Open-iSCSI and IET to cooperate with each other with digests enabled, so I tried with FreeBSD's native target, which "just worked" - well, at least until I launched the second DomU ;) The logs are in "DomU-A-FreeBSD10.log" attached in one of my previous posts.

If I understand correctly, the iSCSI-related issues were only symptoms caused by something Roger seems to have found.

I'm very grateful to you guys for your time and effort to solve my little problem and (especially) for making Xen what it is today.

Kuba

_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxx
http://lists.xen.org/xen-users


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.