On Fri, Jun 04, 2010 at 02:03:26PM +1000, Virgil wrote: > Is this what is happening?: > > http://lists.xensource.com/archives/html/xen-devel/2008-12/msg01030.html > > i.e. the guest does not have a balloon driver causing the PoD cache to run out > of pages and so crashing the domain? > > also: > > 09 - xend integration. > + Always calls xc_hvm_build_target_mem() with memsize=maxmem and > target=memory. If these the same, the internal function will not use > PoD. > > Trying this now. > > Success!! The 32hv12 machine is alive and kicking!!! > > Just set MAXMEM=MEMORY and PoD is disabled!! > > Very happy here. > > Now for WinXP. > Oh, good, you figure it out on your own aswell :) Just write to xen-devel about the *host*/Xen crashes, those sound bad.. Pasi > Cheers > V > > > On Fri, 4 Jun 2010 10:53:40 am Virgil wrote: > > Further notes on WinXP crashes: > > > > What we did was attempt to install XP many many times. Generally what > > seemed to happen was the host machine rebooted when the XP machine > > rebooted. Sometime the XP install rebooted at what seemed the correct > > time. Other times it rebooted for no apparent reason. Sometime the WinXP > > install screen came up other times the host machine was dead, other times > > the host rebooted. > > > > Then I retrieved an old qemu XP image. I simply booted it directly in Xen > > and it started up. After a while it rebooted and took out the host. All > > this was on the xendom0-97/Xen-4.0.0 combo. > > > > I think that when the bad damage occurred to the host. Full rebuild > > yesterday. > > > > Next I tried to hardware virtualize an install on FC12 32 on the > > xendom0-105/Xen-3.9.3 combo. More or less the same thing happened with the > > rebooting. Actually it doesn't seem random..... it always reboots when > > anaconda installs glibc (package number 97). > > > > However! the host never rebooted and still appears quite stable. Haven't > > tried the WinXP on this combo yet. > > > > name="32hv12" > > maxmem="2048" > > memory="1024" > > kernel="/usr/lib/xen/boot/hvmloader" > > builder="hvm" > > device_model="/usr/lib64/xen/bin/qemu-dm" > > disk=[ 'phy:/dev/VolGroup00/lv32HvFc12Spinner,ioemu:xvda,w', > > 'file:/iso/Fedora-12-i386-DVD.iso,ioemu:xvdc:cdrom,r' > > ] > > vif=[ 'type=ioemu, bridge=eth0', ] > > boot="d" > > sdl=0 > > vnc=1 > > vncviewer=1 > > apic=0 > > acpi=0 > > pae=1 > > on_reboot='destroy' > > > > Some of the issue seems to be VNC related. Tried with apic and acpi > > combinations but no joy. Got better screen response by using VNC directly > > on port 5900 to the host rather than using Virtual Machine Manager mostly > > because you can turn off the mouse (possibly some TCP comms issue > > happening to the host - which is on the same LAN). Generally don't know > > what's happening though. > > > > Just looked in xm dmesg now (seems informative - what's a > > "populate-on-demand memory". I don't have enough of it): > > > > (XEN) pg error: p2m_set_entry(): configure P2M table 4KB L2 entry with > > large page (XEN) domain_crash called from p2m.c:1179 > > (XEN) pg error: p2m_set_entry(): configure P2M table 4KB L2 entry with > > large page (XEN) domain_crash called from p2m.c:1179 > > (XEN) pg error: p2m_set_entry(): configure P2M table 4KB L2 entry with > > large page (XEN) domain_crash called from p2m.c:1179 > > (XEN) pg error: p2m_set_entry(): configure P2M table 4KB L2 entry with > > large page (XEN) domain_crash called from p2m.c:1179 > > (XEN) pg error: p2m_set_entry(): configure P2M table 4KB L2 entry with > > large page (XEN) domain_crash called from p2m.c:1179 > > (XEN) pg error: p2m_set_entry(): configure P2M table 4KB L2 entry with > > large page (XEN) domain_crash called from p2m.c:1179 > > > > Lots more of these..... then > > > > (XEN) Domain 9 (vcpu#0) crashed on cpu#0: > > (XEN) ----[ Xen-3.4.3 x86_64 debug=n Not tainted ]---- > > (XEN) CPU: 0 > > (XEN) RIP: 0060:[<00000000c059a8cc>] > > (XEN) RFLAGS: 0000000000200246 CONTEXT: hvm guest > > (XEN) rax: 00000000fffb2000 rbx: 0000000000000000 rcx: 0000000000000400 > > (XEN) rdx: 00000000f47a35c0 rsi: 000000000b8115c0 rdi: 00000000fffb2000 > > (XEN) rbp: 00000000c15cdd74 rsp: 00000000c15cdd68 r8: 0000000000000000 > > (XEN) r9: 0000000000000000 r10: 0000000000000000 r11: 0000000000000000 > > (XEN) r12: 0000000000000000 r13: 0000000000000000 r14: 0000000000000000 > > (XEN) r15: 0000000000000000 cr0: 0000000080050033 cr4: 00000000000006d0 > > (XEN) cr3: 00000000015ac000 cr2: 00000000b3f67000 > > (XEN) ds: 007b es: 007b fs: 00d8 gs: 00e0 ss: 0068 cs: 0060 > > (XEN) pg error: p2m_set_entry(): configure P2M table 4KB L2 entry with > > large page (XEN) domain_crash called from p2m.c:1179 > > (XEN) pg error: p2m_set_entry(): configure P2M table 4KB L2 entry with > > large page (XEN) domain_crash called from p2m.c:1179 > > (XEN) pg error: p2m_set_entry(): configure P2M table 4KB L2 entry with > > large page (XEN) domain_crash called from p2m.c:1179 > > (XEN) p2m_pod_demand_populate: Out of populate-on-demand memory! tot_pages > > 262343 pod_entries 262944 (XEN) domain_crash called from p2m.c:1065 > > (XEN) Domain 12 (vcpu#0) crashed on cpu#0: > > (XEN) ----[ Xen-3.4.3 x86_64 debug=n Not tainted ]---- > > (XEN) CPU: 0 > > (XEN) RIP: 0060:[<00000000c059a8cc>] > > (XEN) RFLAGS: 0000000000000246 CONTEXT: hvm guest > > (XEN) rax: 00000000fffb2000 rbx: 0000000000000000 rcx: 0000000000000400 > > (XEN) rdx: 000000004e4ed000 rsi: 00000000b1b5f000 rdi: 00000000fffb2000 > > (XEN) rbp: 00000000c15a7d74 rsp: 00000000c15a7d68 r8: 0000000000000000 > > (XEN) r9: 0000000000000000 r10: 0000000000000000 r11: 0000000000000000 > > (XEN) r12: 0000000000000000 r13: 0000000000000000 r14: 0000000000000000 > > (XEN) r15: 0000000000000000 cr0: 000000008005003b cr4: 00000000000006d0 > > (XEN) cr3: 00000000015a9000 cr2: 00000000b1b60000 > > (XEN) ds: 007b es: 007b fs: 00d8 gs: 00e0 ss: 0068 cs: 0060 > > > > On Thu, 3 Jun 2010 04:59:12 pm you wrote: > > > On Thu, Jun 03, 2010 at 02:23:15PM +1000, Virgil wrote: > > > > On Thu, 3 Jun 2010 05:34:34 am M A Young wrote: > > > > > On Wed, 2 Jun 2010, Virgil wrote: > > > > > > Just doing an rpmbuild on: > > > > > > > > > > > > xen-4.0.0-0.7.fc12.x86_64 > > > > > > > > > > > > and I get this error towards the end: > > > > > > > > > > > > Processing files: xen-debuginfo-4.0.0-0.7.fc12.x86_64 > > > > > > Checking for unpackaged file(s): /usr/lib/rpm/check-files > > > > > > /root/rpmbuild/BUILDROOT/xen-4.0.0-0.7.fc12.x86_64 > > > > > > > > > > > > error: Installed (but unpackaged) file(s) found: > > > > > > /etc/hotplug/xen-backend.agent > > > > > > > > > > If you actually got some RPMs built then you should be able to use > > > > > them. It looks like you might have the hotplug package installed > > > > > (current Fedora doesn't have this package) and xen is trying to set > > > > > things up accordingly. > > > > > > > > > > If you didn't get any RPMs I did a test build of xen-4.0.0-0.7.fc12 > > > > > at http://koji.fedoraproject.org/koji/taskinfo?taskID=2225738 for > > > > > rawhide that you might be able to use. > > > > > > > > > > Michael Young > > > > > > > > OK have rebuilt the machine from the ground up. DVD install then `yum > > > > update` > > > > > > > > What works: > > > > > > > > xendom0-97/xen-4.0.0 runs most domU images (not WinXP - runs for a > > > > while then locks up). Even the old FC6 domUs run well. > > > > > > Anything in "xm log" when the WinXP locks up? or "xm dmesg" ? or in > > > /var/log/xen/* > > > > > > > What doesn't: > > > > > > > > xendom0-105/xen-4.0.0 crashes and burns badly with a pv32fc12 domU (so > > > > badly the keyboard lights blink in pain. Reset button required. Lots > > > > and lots of hypervisorish stuff all over the screen. Very likely the > > > > cause is noted in there somewhere). > > > > > > Please set up a serial console so you can capture the full boot and > > > error/crash messages: http://wiki.xensource.com/xenwiki/XenSerialConsole > > > > > > A couple of questions: > > > > > > - Do you have "nomodeset" parameter for the dom0 kernel? > > > - Did you limit dom0 memory to, say, 1024 MB ? > > > > > > http://wiki.xensource.com/xenwiki/XenBestPractices > > > > > > > Not tested: > > > > xen-3.4.3 with either kernel. Is this version of Xen more likely to > > > > react in a more stable manner? Or is this a kernel thing? > > > > > > You can try. I'd guess it's the dom0 kernel, and not Xen itself. > > > > > > -- Pasi > > > > > > > ????????? Xen-fc12 > > > > > > > > ????????? xen-3.4.3 > > > > ??? ????????? xen-3.4.3-2.fc12.x86_64.rpm > > > > ??? ????????? xen-devel-3.4.3-2.fc12.x86_64.rpm > > > > ??? ????????? xen-doc-3.4.3-2.fc12.x86_64.rpm > > > > ??? ????????? xen-hypervisor-3.4.3-2.fc12.x86_64.rpm > > > > ??? ????????? xen-libs-3.4.3-2.fc12.x86_64.rpm > > > > ??? ????????? xen-runtime-3.4.3-2.fc12.x86_64.rpm > > > > ????????? xen-4.0.0 > > > > ??? ????????? xen-4.0.0-0.7.fc12.src.rpm > > > > ??? ????????? xen-4.0.0-0.7.fc12.x86_64.rpm > > > > ??? ????????? xen-debuginfo-4.0.0-0.7.fc12.x86_64.rpm > > > > ??? ????????? xen-devel-4.0.0-0.7.fc12.x86_64.rpm > > > > ??? ????????? xen-doc-4.0.0-0.7.fc12.x86_64.rpm > > > > ??? ????????? xen-hypervisor-4.0.0-0.7.fc12.x86_64.rpm > > > > ??? ????????? xen-libs-4.0.0-0.7.fc12.x86_64.rpm > > > > ??? ????????? xen-runtime-4.0.0-0.7.fc12.x86_64.rpm > > > > ????????? xendom0-105 > > > > ??? ????????? kernel-2.6.32.14-1.2.105.xendom0.fc12.x86_64.rpm > > > > ??? ????????? > > > > kernel-devel-2.6.32.14-1.2.105.xendom0.fc12.x86_64.rpm ??? > > > > ????????? kernel-firmware-2.6.32.14-1.2.105.xendom0.fc12.noarch.rpm > > > > ??? ????????? > > > > kernel-headers-2.6.32.14-1.2.105.xendom0.fc12.x86_64.rpm ????????? > > > > xendom0-97 > > > > > > > > ????????? kernel-2.6.32.11-1.2.97.xendom0.fc12.x86_64.rpm > > > > ????????? kernel-devel-2.6.32.11-1.2.97.xendom0.fc12.x86_64.rpm > > > > ????????? > > > > kernel-firmware-2.6.32.11-1.2.97.xendom0.fc12.noarch.rpm > > > > ????????? > > > > kernel-headers-2.6.32.11-1.2.97.xendom0.fc12.x86_64.rpm > > > > > > > > Cheers > > > > V > > > > -- > > > > xen mailing list > > > > xen@xxxxxxxxxxxxxxxxxxxxxxx > > > > https://admin.fedoraproject.org/mailman/listinfo/xen > > > > -- > > xen mailing list > > xen@xxxxxxxxxxxxxxxxxxxxxxx > > https://admin.fedoraproject.org/mailman/listinfo/xen -- xen mailing list xen@xxxxxxxxxxxxxxxxxxxxxxx https://admin.fedoraproject.org/mailman/listinfo/xen
