WARNING - OLD ARCHIVES

This is an archived copy of the Xen.org mailing list, which we have preserved to ensure that existing links to archives are not broken. The live archive, which contains the latest emails, can be found at http://lists.xen.org/
   
 
 
Xen 
 
Home Products Support Community News
 
   
 

xen-devel

[Xen-devel] HVM save/restore testing report in c/s 13788

To: <xen-devel@xxxxxxxxxxxxxxxxxxx>
Subject: [Xen-devel] HVM save/restore testing report in c/s 13788
From: "Yu, Ping Y" <ping.y.yu@xxxxxxxxx>
Date: Wed, 7 Feb 2007 17:37:10 +0800
Delivery-date: Wed, 07 Feb 2007 01:38:01 -0800
Envelope-to: www-data@xxxxxxxxxxxxxxxxxx
List-help: <mailto:xen-devel-request@lists.xensource.com?subject=help>
List-id: Xen developer discussion <xen-devel.lists.xensource.com>
List-post: <mailto:xen-devel@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
Sender: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
Thread-index: AcdKm4owvtY5CnhSRlyukDkRCD6gFQ==
Thread-topic: HVM save/restore testing report in c/s 13788
Hi, all

We have performed save/restore testing in c/s 13778. In this testing, we
just focus on basic save/restore function in multiple arch/os/cpu
combinations, and the total combination is
6(service/os)*2(linux/windows)*2(OS UP/SMP)*3(vcpus 1/2/4)=72.
In the testing, we found that the basic function is OK in most
platforms, but sometimes, "xm save" might fail occasionally. 

Here are the bugs we found in the testing.
(1) [SR]In x86_64, after "xm save", xm  command is abnormal
http://bugzilla.xensource.com/bugzilla/show_bug.cgi?id=885
This should be a P1 bug, for it will break all xm command after "xm
save"

(2) [SR]save a 6G memory guest failed
http://bugzilla.xensource.com/bugzilla/show_bug.cgi?id=890
6G memory is frequently used in PAE, so this is a P1 bug

(3) [SR]the restored guest could not save again in x86_64
http://bugzilla.xensource.com/bugzilla/show_bug.cgi?id=887
HVM could be saved multiple times, so it is P1 bug

(4) [SR]save fails for HVM with vcpus=8
http://bugzilla.xensource.com/bugzilla/show_bug.cgi?id=889
This bug is caused by insufficient space for too much vcpus, and should
be a P1 bug

(5) [SR]if there is no sufficient disk space, "xm save" will fail
http://bugzilla.xensource.com/bugzilla/show_bug.cgi?id=888
This bug is a user experience bug, and set it as P2

(6) [SR]save HVM guest might fail in multiple scenario
http://bugzilla.xensource.com/bugzilla/show_bug.cgi?id=884
This bug is not easy to produce, and maybe it is related with bug 888,
so I think it is P2 bug

(7) [SR]restore SMP guest in IA32e will show some trace info
http://bugzilla.xensource.com/bugzilla/show_bug.cgi?id=886
There are many trace info in the restored guest, and should be a P2 bug

(8) [SR]"xm save" will overwirte the disk image without any notification
http://bugzilla.xensource.com/bugzilla/show_bug.cgi?id=891
A user experience bug, and set as P2


Here is the detail report for each combination, here "OK" means the
basic function, such as keyboard,mouse,network,console is OK, and
"Unstable" means that sometimes basic function is not good, and we have
bug 884 (http://bugzilla.xensource.com/bugzilla/show_bug.cgi?id=884 )for
that. 

Ia32-on-IA32
========================
                  vcpu=1        vcpu=2  vcpu=4
Linux_UP                OK      OK              OK
Linux_SMP               OK      OK              OK
Windows_UP         OK   OK              OK
Windows_SMP        OK   OK              OK

IA32-on-PAE
========================
                  vcpu=1        vcpu=2  vcpu=4
Linux_UP                OK      OK              OK
Linux_SMP               OK      OK              OK
Windows_UP         OK   OK              OK
Windows_SMP        OK   OK              OK

PAE-on-PAE
========================
                  vcpu=1        vcpu=2  vcpu=4
Linux_UP                OK      OK              OK
Linux_SMP               OK      OK              OK
Windows_UP         OK   OK              OK
Windows_SMP        OK   OK              OK

IA32-on-x86_64
========================
                  vcpu=1        vcpu=2  vcpu=4
Linux_UP                OK      Unstable        OK
Linux_SMP               OK      Unstable        OK
Windows_UP         OK   OK              OK
Windows_SMP        OK   OK              OK

PAE-on-x86_64
========================
                  vcpu=1        vcpu=2  vcpu=4
Linux_UP                OK      OK              OK
Linux_SMP               OK      OK              OK
Windows_UP         OK   OK              OK
Windows_SMP        OK   OK              OK

x86_64-on-x86_64
========================
                  vcpu=1                vcpu=2  vcpu=4
Linux_UP                Unstable        OK              OK
Linux_SMP               Unstable        Unstable        Unstable
Windows_UP         OK           OK              OK
Windows_SMP        OK           OK              OK



-- Ping

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel

<Prev in Thread] Current Thread [Next in Thread>
  • [Xen-devel] HVM save/restore testing report in c/s 13788, Yu, Ping Y <=