I have built a new kernel (kernel-2.6.32.21-167.xendom0.fc12) at http://koji.fedoraproject.org/koji/taskinfo?taskID=2445999 or at the repository http://repos.fedorapeople.org/repos/myoung/dom0-kernel/
This includes the net-sched-plug feature built as a module.
Michael Young
I have built a new kernel (kernel-2.6.32.21-168.xendom0.fc12) at http://koji.fedoraproject.org/koji/taskinfo?taskID=2477887 or at the repository http://repos.fedorapeople.org/repos/myoung/dom0-kernel/ It includes fixes for the recent x86_64 local root exploit.
There are also some new builds of xen. The Fedora packages xen-4.0.1-3.fc15 and xen-4.0.1-3.fc14 available via http://koji.fedoraproject.org/koji/packageinfo?packageID=7 have some IRQ fixes that solve a problem I was having with a keyboard and mouse and the patch that disables xsave which was causing problems with HVM (both these are from 4.0.2-rc1-pre). It also creates a symbolic link for qemu-dm from the place it was in 3.4.x to where it is in 4.0.x for compatibility.
I also have a temporary test build xen-4.0.1-1.fc13.1 for Fedora 13 at http://koji.fedoraproject.org/koji/taskinfo?taskID=2478501 . This is slightly misnamed as it is actually of 4.0.2-rc1-pre and includes the fixes above. Note that he build system will delete those RPMs automatically after a week or so.
Michael Young
On 09/24/2010 06:02 PM, M A Young wrote:
I have built a new kernel (kernel-2.6.32.21-168.xendom0.fc12) at http://koji.fedoraproject.org/koji/taskinfo?taskID=2477887 or at the repositoryhttp://repos.fedorapeople.org/repos/myoung/dom0-kernel/ It includes fixes for the recent x86_64 local root exploit.
I also have a temporary test build xen-4.0.1-1.fc13.1 for Fedora 13 at http://koji.fedoraproject.org/koji/taskinfo?taskID=2478501 . This is slightly misnamed as it is actually of 4.0.2-rc1-pre and includes the fixes above. Note that he build system will delete those RPMs automatically after a week or so.
I'm trying these two under f13, and not getting past udev starting on the pv_ops kernel. (Xen oops) The non-xen kernel works fine on this hardware. A pv_ops version from back around January was working well for me, but I wanted to grab the x86_64 multilib-problem patched version. That brought me to upgrade my xen-* versions, and f12 to f13.
The top of the stack has 'xen_force_evtchn_callback' on it. I see udev's [OK] message, and then it very reliably crashes (I can get it to tell me it's going down for reboot with a three-finger salute, but nothing really happens).
Is a serial console the only way to get a good Xen log? Looking at this board, I think I need to take the heatsink off the CPU to get at the serial header, so I haven't tried yet...
-Bill
On Sat, 25 Sep 2010, Bill McGonigle wrote:
I'm trying these two under f13, and not getting past udev starting on the pv_ops kernel. (Xen oops) The non-xen kernel works fine on this hardware. A pv_ops version from back around January was working well for me, but I wanted to grab the x86_64 multilib-problem patched version. That brought me to upgrade my xen-* versions, and f12 to f13.
The top of the stack has 'xen_force_evtchn_callback' on it. I see udev's [OK] message, and then it very reliably crashes (I can get it to tell me it's going down for reboot with a three-finger salute, but nothing really happens).
Is a serial console the only way to get a good Xen log? Looking at this board, I think I need to take the heatsink off the CPU to get at the serial header, so I haven't tried yet...
In your case it might be the only way to get the relevant information. I also have a computer without a serial port, and have tried other two things to get debugging information though they may not help in your case.
If the computer boots far enough you can get the xen serial output with the xm dmesg or xl dmesg commands. Adding the console_to_ring and conring_size=65536 options to the xen boot line may give more information.
I have also booted a customized xen dom0 live CD as a KVM guest where you can redirect the console output to a file in the parent machine. That does have the disadvantage that some of the hardware it sees is emulated so it may behave differently.
If you haven't already seen them it is worth looking at http://wiki.xen.org/xenwiki/XenParavirtOps http://wiki.xen.org/xenwiki/XenSerialConsole
Michael Young
On Sat, 25 Sep 2010, Bill McGonigle wrote:
I'm trying these two under f13, and not getting past udev starting on the pv_ops kernel. (Xen oops) The non-xen kernel works fine on this hardware. A pv_ops version from back around January was working well for me, but I wanted to grab the x86_64 multilib-problem patched version. That brought me to upgrade my xen-* versions, and f12 to f13.
The top of the stack has 'xen_force_evtchn_callback' on it. I see udev's [OK] message, and then it very reliably crashes (I can get it to tell me it's going down for reboot with a three-finger salute, but nothing really happens).
I had a further thought - can you run rpm -V xen-runtime and see if it reports anything. In particular there are two files /etc/udev/rules.d/xen-backend.rules and /etc/udev/rules.d/xend.rules that are related to udev setup of xen and if you modified them prior to the update then you might have an out-of-date configuration.
Michael Young
On 09/25/2010 09:48 AM, M A Young wrote:
In your case it might be the only way to get the relevant information. I also have a computer without a serial port, and have tried other two things to get debugging information though they may not help in your case.
OK, after much futzing around, I realized the IDC10->DB9M adapters I had on-hand are all improperly wired ('DTK-type' vs. needed 'AT-Everest' type) for this mobo. I put in an order for a new one and will get back to it in a couple days (unless I get the soldering iron out first...).
I had a further thought - can you run rpm -V xen-runtime and see if it reports anything. In particular there are two files /etc/udev/rules.d/xen-backend.rules and /etc/udev/rules.d/xend.rules that are related to udev setup of xen and if you modified them prior to the update then you might have an out-of-date configuration.
Ah, good thought - that was true but fixing it to be current with xen-runtime-4.0.1-1.fc13.1.x86_64 didn't help this problem. I thought that if udev/rules.d/xen* was causing a problem, removing those files might let the boot succeed. It didn't. I don't know if that means anything or not.
I'll post the log here asap.
Thanks, -Bill
On 09/25/2010 09:48 AM, M A Young wrote:
In your case it might be the only way to get the relevant information.
OK, new backplate works like a charm, appending the serial log.
Now that the messages aren't mostly scrolled off the screen (can I get a better VGA mode than 80x25 under Xen?), I saw more information which led me to find this thread:
http://lists.xensource.com/archives/html/xen-devel/2010-09/msg01482.html
In my case I don't really want to constrain dom0's memory on this machine, yet (it's still running services that ideally would be moved to a DomU but the machine is a bit core-starved).
This is an x86_64 dom0. grub for this boot is:
kernel /xen.gz dom0_max_vcpus=1 dom0_vcpus_pin loglvl=all guest_loglvl=all com1=9600,8n1 sync_console console_to_ring conring_size=65536 console=com1 module /vmlinuz-2.6.32.21-168.xendom0.fc12.x86_64 ro root=LABEL=librescu-root selinux=0 SYSFONT=latarcyrheb-sun16 LANG=en_US.UTF-8 KEYTABLE=us rootflags=data=journal acpi=off earlyprintk=xen console=hvc0 nomodeset module /initramfs-2.6.32.21-168.xendom0.fc12.x86_64.img
-Bill
Quick question:
Should the libvirt stuff be updated?
Sep 27 10:47:41 seanl64 libvirtd: 10:47:41.234: error : statsErrorFunc:60 : internal error read_bd_stats: Failed to read any block statistics
libvirt-client-0.8.2-1.fc12.x86_64 libvirt-python-0.8.2-1.fc12.x86_64 libvirt-0.8.2-1.fc12.x86_64
xen-runtime-4.0.1-1.fc12.1.x86_64 xen-devel-4.0.1-1.fc12.1.x86_64 xen-libs-4.0.1-1.fc12.1.x86_64 xen-debuginfo-4.0.1-1.fc12.1.x86_64 xen-4.0.1-1.fc12.1.x86_64 xen-doc-4.0.1-1.fc12.1.x86_64 xen-hypervisor-4.0.1-1.fc12.1.x86_64
Linux seanl64.xxxxxx.xx.xx 2.6.32.21-168.xendom0.fc12.x86_64 #1 SMP Mon Sep 20 19:32:37 UTC 2010 x86_64 x86_64 x86_64 GNU/Linux
Cheers Virgil
On Sat, 25 Sep 2010 08:02:00 am M A Young wrote:
I have built a new kernel (kernel-2.6.32.21-168.xendom0.fc12) at http://koji.fedoraproject.org/koji/taskinfo?taskID=2477887 or at the repository http://repos.fedorapeople.org/repos/myoung/dom0-kernel/ It includes fixes for the recent x86_64 local root exploit.
There are also some new builds of xen. The Fedora packages xen-4.0.1-3.fc15 and xen-4.0.1-3.fc14 available via http://koji.fedoraproject.org/koji/packageinfo?packageID=7 have some IRQ fixes that solve a problem I was having with a keyboard and mouse and the patch that disables xsave which was causing problems with HVM (both these are from 4.0.2-rc1-pre). It also creates a symbolic link for qemu-dm from the place it was in 3.4.x to where it is in 4.0.x for compatibility.
I also have a temporary test build xen-4.0.1-1.fc13.1 for Fedora 13 at http://koji.fedoraproject.org/koji/taskinfo?taskID=2478501 . This is slightly misnamed as it is actually of 4.0.2-rc1-pre and includes the fixes above. Note that he build system will delete those RPMs automatically after a week or so.
Michael Young
xen mailing list xen@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/xen
Further to the previous email regarding diskio stats:
It appears the ksysguardd is no longer able to gather stats either.
I'm suspecting a bug in the kernel and not libvirtd.
Cheers V
On Sat, 25 Sep 2010 08:02:00 am M A Young wrote:
I have built a new kernel (kernel-2.6.32.21-168.xendom0.fc12) at http://koji.fedoraproject.org/koji/taskinfo?taskID=2477887 or at the repository http://repos.fedorapeople.org/repos/myoung/dom0-kernel/ It includes fixes for the recent x86_64 local root exploit.
There are also some new builds of xen. The Fedora packages xen-4.0.1-3.fc15 and xen-4.0.1-3.fc14 available via http://koji.fedoraproject.org/koji/packageinfo?packageID=7 have some IRQ fixes that solve a problem I was having with a keyboard and mouse and the patch that disables xsave which was causing problems with HVM (both these are from 4.0.2-rc1-pre). It also creates a symbolic link for qemu-dm from the place it was in 3.4.x to where it is in 4.0.x for compatibility.
I also have a temporary test build xen-4.0.1-1.fc13.1 for Fedora 13 at http://koji.fedoraproject.org/koji/taskinfo?taskID=2478501 . This is slightly misnamed as it is actually of 4.0.2-rc1-pre and includes the fixes above. Note that he build system will delete those RPMs automatically after a week or so.
Michael Young
xen mailing list xen@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/xen
On Fri, Sep 24, 2010 at 11:02:00PM +0100, M A Young wrote:
I have built a new kernel (kernel-2.6.32.21-168.xendom0.fc12) at http://koji.fedoraproject.org/koji/taskinfo?taskID=2477887 or at the repository http://repos.fedorapeople.org/repos/myoung/dom0-kernel/ It includes fixes for the recent x86_64 local root exploit.
There are also some new builds of xen. The Fedora packages xen-4.0.1-3.fc15 and xen-4.0.1-3.fc14 available via http://koji.fedoraproject.org/koji/packageinfo?packageID=7 have some IRQ fixes that solve a problem I was having with a keyboard and mouse and the patch that disables xsave which was causing problems with HVM (both these are from 4.0.2-rc1-pre). It also creates a symbolic link for qemu-dm from the place it was in 3.4.x to where it is in 4.0.x for compatibility.
Are you planning to get xen-4.0.1-3.fc14 to F14 before it goes GA? That would be good..
It seems the current F14 tree still has xen-4.0.1-1.fc14.src.rpm.
I also have a temporary test build xen-4.0.1-1.fc13.1 for Fedora 13 at http://koji.fedoraproject.org/koji/taskinfo?taskID=2478501 . This is slightly misnamed as it is actually of 4.0.2-rc1-pre and includes the fixes above. Note that he build system will delete those RPMs automatically after a week or so.
Could you upload xen-4.0.1-1.fc13.1 src.rpm somewhere? I'd like to mirror it, some people might want to get it also now when it's removed from that koji page..
-- Pasi
Are you planning to get xen-4.0.1-3.fc14 to F14 before it goes GA? That would be good..
The most recent "yum update" installs xen-4.0.1-4.fc14 on F14 ( rawhide)
Boris.
--- On Sun, 10/3/10, Pasi Kärkkäinen pasik@iki.fi wrote:
From: Pasi Kärkkäinen pasik@iki.fi Subject: Re: [Fedora-xen] xen and dom0 kernel builds To: "M A Young" m.a.young@durham.ac.uk Cc: xen@lists.fedoraproject.org Date: Sunday, October 3, 2010, 11:59 AM
On Fri, Sep 24, 2010 at 11:02:00PM +0100, M A Young wrote:
I have built a new kernel (kernel-2.6.32.21-168.xendom0.fc12) at http://koji.fedoraproject.org/koji/taskinfo?taskID=2477887 or at the repository http://repos.fedorapeople.org/repos/myoung/dom0-kernel/ It includes fixes for the recent x86_64 local root exploit.
There are also some new builds of xen. The Fedora packages xen-4.0.1-3.fc15 and xen-4.0.1-3.fc14 available via http://koji.fedoraproject.org/koji/packageinfo?packageID=7 have some IRQ fixes that solve a problem I was having with a keyboard and mouse and the patch that disables xsave which was causing problems with HVM (both these are from 4.0.2-rc1-pre). It also creates a symbolic link for qemu-dm from the place it was in 3.4.x to where it is in 4.0.x for compatibility.
Are you planning to get xen-4.0.1-3.fc14 to F14 before it goes GA? That would be good..
It seems the current F14 tree still has xen-4.0.1-1.fc14.src.rpm.
I also have a temporary test build xen-4.0.1-1.fc13.1 for Fedora 13 at http://koji.fedoraproject.org/koji/taskinfo?taskID=2478501 . This is slightly misnamed as it is actually of 4.0.2-rc1-pre and includes the fixes above. Note that he build system will delete those RPMs automatically after a week or so.
Could you upload xen-4.0.1-1.fc13.1 src.rpm somewhere? I'd like to mirror it, some people might want to get it also now when it's removed from that koji page..
-- Pasi
-- xen mailing list xen@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/xen
On Sun, Oct 03, 2010 at 10:46:49AM -0700, Boris Derzhavets wrote:
Are you planning to get xen-4.0.1-3.fc14 to F14 before it goes GA? That would be good..
The most recent "yum update" installs xen-4.0.1-4.fc14 on F14 ( rawhide)
I think rawhide is currently F15.. F14 has been branched already.
But yeah, I re-checked, and I noticed xen-4.0.1-4.fc14.src.rpm in F14 updates-testing.
So all fine.
Thanks!
-- Pasi
Boris.
--- On Sun, 10/3/10, Pasi Kärkkäinen pasik@iki.fi wrote:
From: Pasi Kärkkäinen <pasik@iki.fi> Subject: Re: [Fedora-xen] xen and dom0 kernel builds To: "M A Young" <m.a.young@durham.ac.uk> Cc: xen@lists.fedoraproject.org Date: Sunday, October 3, 2010, 11:59 AM On Fri, Sep 24, 2010 at 11:02:00PM +0100, M A Young wrote: > I have built a new kernel (kernel-2.6.32.21-168.xendom0.fc12) at > [1]http://koji.fedoraproject.org/koji/taskinfo?taskID=2477887 or at the > repository [2]http://repos.fedorapeople.org/repos/myoung/dom0-kernel/ > It includes fixes for the recent x86_64 local root exploit. > > There are also some new builds of xen. The Fedora packages > xen-4.0.1-3.fc15 and xen-4.0.1-3.fc14 available via > [3]http://koji.fedoraproject.org/koji/packageinfo?packageID=7 have some IRQ > fixes that solve a problem I was having with a keyboard and mouse and the > patch that disables xsave which was causing problems with HVM (both these > are from 4.0.2-rc1-pre). It also creates a symbolic link for qemu-dm from > the place it was in 3.4.x to where it is in 4.0.x for compatibility. > Are you planning to get xen-4.0.1-3.fc14 to F14 before it goes GA? That would be good.. It seems the current F14 tree still has xen-4.0.1-1.fc14.src.rpm. > I also have a temporary test build xen-4.0.1-1.fc13.1 for Fedora 13 at > [4]http://koji.fedoraproject.org/koji/taskinfo?taskID=2478501 . This is > slightly misnamed as it is actually of 4.0.2-rc1-pre and includes the > fixes above. Note that he build system will delete those RPMs > automatically after a week or so. > Could you upload xen-4.0.1-1.fc13.1 src.rpm somewhere? I'd like to mirror it, some people might want to get it also now when it's removed from that koji page.. -- Pasi -- xen mailing list [5]xen@lists.fedoraproject.org [6]https://admin.fedoraproject.org/mailman/listinfo/xen
References
Visible links
- http://koji.fedoraproject.org/koji/taskinfo?taskID=2477887
- http://repos.fedorapeople.org/repos/myoung/dom0-kernel/
- http://koji.fedoraproject.org/koji/packageinfo?packageID=7
- http://koji.fedoraproject.org/koji/taskinfo?taskID=2478501
- file:///mc/compose?to=xen@lists.fedoraproject.org
- https://admin.fedoraproject.org/mailman/listinfo/xen
On Sun, 3 Oct 2010, Pasi Kärkkäinen wrote:
Are you planning to get xen-4.0.1-3.fc14 to F14 before it goes GA? That would be good..
There has been a bit of a delay because it got built with a bad gcc. xen-4.0.1-4.fc14 is a rebuild once gcc was fixed. As specified in https://fedoraproject.org/wiki/Updates_Policy I am now waiting a week before I push it to stable.
Could you upload xen-4.0.1-1.fc13.1 src.rpm somewhere? I'd like to mirror it, some people might want to get it also now when it's removed from that koji page..
I could but I have now just built xen-4.0.1-1.fc13.2 at http://koji.fedoraproject.org/koji/taskinfo?taskID=2509285 This is the same except that I am experimenting by adding a patch from the Fedora grub package to see if I can get ext4 support in pvgrub.
Michael Young
On Sun, Oct 03, 2010 at 08:55:15PM +0100, M A Young wrote:
On Sun, 3 Oct 2010, Pasi Kärkkäinen wrote:
Are you planning to get xen-4.0.1-3.fc14 to F14 before it goes GA? That would be good..
There has been a bit of a delay because it got built with a bad gcc. xen-4.0.1-4.fc14 is a rebuild once gcc was fixed. As specified in https://fedoraproject.org/wiki/Updates_Policy I am now waiting a week before I push it to stable.
Ok.
Could you upload xen-4.0.1-1.fc13.1 src.rpm somewhere? I'd like to mirror it, some people might want to get it also now when it's removed from that koji page..
I could but I have now just built xen-4.0.1-1.fc13.2 at http://koji.fedoraproject.org/koji/taskinfo?taskID=2509285 This is the same except that I am experimenting by adding a patch from the Fedora grub package to see if I can get ext4 support in pvgrub.
Great. Thanks!
-- Pasi
On Sun, 3 Oct 2010, M A Young wrote:
On Sun, 3 Oct 2010, Pasi Kärkkäinen wrote:
There has been a bit of a delay because it got built with a bad gcc. xen-4.0.1-4.fc14 is a rebuild once gcc was fixed. As specified in https://fedoraproject.org/wiki/Updates_Policy I am now waiting a week before I push it to stable.
xen-4.0.1-4.fc14 has now made it into the F14 release. I have also now built xen-4.0.1-5.fc14/15 which adds the fix for ext4 in pvgrub I was testing.
I have also done a new kernel which is at http://koji.fedoraproject.org/koji/taskinfo?taskID=2518648 or the repository http://repos.fedorapeople.org/repos/myoung/dom0-kernel/fedora-12/ with the latest xen/stable-2.6.32.x patches including the watchdog driver (built as a module) and some IRQ and memory fixes.
Michael Young
On Fri, Oct 08, 2010 at 09:56:35PM +0100, M A Young wrote:
On Sun, 3 Oct 2010, M A Young wrote:
On Sun, 3 Oct 2010, Pasi Kärkkäinen wrote:
There has been a bit of a delay because it got built with a bad gcc. xen-4.0.1-4.fc14 is a rebuild once gcc was fixed. As specified in https://fedoraproject.org/wiki/Updates_Policy I am now waiting a week before I push it to stable.
xen-4.0.1-4.fc14 has now made it into the F14 release. I have also now built xen-4.0.1-5.fc14/15 which adds the fix for ext4 in pvgrub I was testing.
I have also done a new kernel which is at http://koji.fedoraproject.org/koji/taskinfo?taskID=2518648 or the repository http://repos.fedorapeople.org/repos/myoung/dom0-kernel/fedora-12/ with the latest xen/stable-2.6.32.x patches including the watchdog driver (built as a module) and some IRQ and memory fixes.
Thanks, good work!
-- Pasi
I have now built xen-4.0.1-6.fc14/5 to fix a boot failure for some hardware ( https://bugzilla.redhat.com/show_bug.cgi?id=642108 ) available from koji at http://koji.fedoraproject.org/koji/packageinfo?packageID=7 and also a test build xen-4.0.1-1.fc13.3 built for f13 temporarily available at http://koji.fedoraproject.org/koji/taskinfo?taskID=2530814 with the same patch.
I should be able to get xen-4.0.1-5.fc14 into the released version on Fedora 14, but (unless there is enough karma to speed up the process) I believe it is too late for xen-4.0.1-6.fc14 which will probably end up as a 0-day update (unless it has been superceded by then).
Michael Young
How can we increase the karma? I can add some...
Will this xen-4.0.1-1.fc13.3 work for bridged rather than routed? Sorry I am not up on the latest xen stuff still stuck on F8 :(
If you do get it in, what would that mean for dom0 on F14? What would be missing from a usable dom0?
Thanks for your work on this.
On 10/12/10 4:08 PM, M A Young wrote:
I have now built xen-4.0.1-6.fc14/5 to fix a boot failure for some hardware ( https://bugzilla.redhat.com/show_bug.cgi?id=642108 ) available from koji at http://koji.fedoraproject.org/koji/packageinfo?packageID=7 and also a test build xen-4.0.1-1.fc13.3 built for f13 temporarily available at http://koji.fedoraproject.org/koji/taskinfo?taskID=2530814 with the same patch.
I should be able to get xen-4.0.1-5.fc14 into the released version on Fedora 14, but (unless there is enough karma to speed up the process) I believe it is too late for xen-4.0.1-6.fc14 which will probably end up as a 0-day update (unless it has been superceded by then).
Michael Young
xen mailing list xen@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/xen
On Wed, 13 Oct 2010, fcxen user wrote:
How can we increase the karma? I can add some...
You can vote on xen-4.0.1-5.fc14 at https://admin.fedoraproject.org/updates/xen-4.0.1-5.fc14 though you may need to be logged in for your vote to count.
Will this xen-4.0.1-1.fc13.3 work for bridged rather than routed? Sorry I am not up on the latest xen stuff still stuck on F8 :(
I use the virtual bridge provided by the libvirt package. I didn't have a lot of success using a true bridge last time I tried it, but that would have been with xen 3.x not 4.x.
If you do get it in, what would that mean for dom0 on F14? What would be missing from a usable dom0?
I use the F12 kernel I build on F14 which seems to work (I was following F12 because that is still on 2.6.32.x which is has the best maintained dom0 pvops patches at present). It did have trouble with systemd but that was dropped as the default init for F14.
Michael Young
On Wed, 13 Oct 2010, M A Young wrote:
On Wed, 13 Oct 2010, fcxen user wrote:
How can we increase the karma? I can add some...
You can vote on xen-4.0.1-5.fc14 at https://admin.fedoraproject.org/updates/xen-4.0.1-5.fc14 though you may need to be logged in for your vote to count.
xen-4.0.1-5.fc14 got enough karma and made it into the F14 release and I have now submitted xen-4.0.1-6.fc14 for updates-testing.
Please test it (it is available from http://koji.fedoraproject.org/koji/buildinfo?buildID=200238 and will be in updates-testing soon) and consider reporting your results at https://admin.fedoraproject.org/updates/xen-4.0.1-6.fc14 as it may still be possible to get it into the F14 release.
Michael Young
On Thu, Oct 14, 2010 at 04:32:20PM +0100, M A Young wrote:
On Wed, 13 Oct 2010, M A Young wrote:
On Wed, 13 Oct 2010, fcxen user wrote:
How can we increase the karma? I can add some...
You can vote on xen-4.0.1-5.fc14 at https://admin.fedoraproject.org/updates/xen-4.0.1-5.fc14 though you may need to be logged in for your vote to count.
xen-4.0.1-5.fc14 got enough karma and made it into the F14 release and I have now submitted xen-4.0.1-6.fc14 for updates-testing.
Please test it (it is available from http://koji.fedoraproject.org/koji/buildinfo?buildID=200238 and will be in updates-testing soon) and consider reporting your results at https://admin.fedoraproject.org/updates/xen-4.0.1-6.fc14 as it may still be possible to get it into the F14 release.
4.0.1-6.fc14 works for me!
-- Pasi
xen-4.0.1-6.fc12 (recompiled under FC12) likewise seems to resolve all issues here. Fingers crossed, so far so good.
Cheers V
On Sun, 17 Oct 2010 10:32:12 pm Pasi Kärkkäinen wrote:
On Thu, Oct 14, 2010 at 04:32:20PM +0100, M A Young wrote:
On Wed, 13 Oct 2010, M A Young wrote:
On Wed, 13 Oct 2010, fcxen user wrote:
How can we increase the karma? I can add some...
You can vote on xen-4.0.1-5.fc14 at https://admin.fedoraproject.org/updates/xen-4.0.1-5.fc14 though you may need to be logged in for your vote to count.
xen-4.0.1-5.fc14 got enough karma and made it into the F14 release and I have now submitted xen-4.0.1-6.fc14 for updates-testing.
Please test it (it is available from http://koji.fedoraproject.org/koji/buildinfo?buildID=200238 and will be in updates-testing soon) and consider reporting your results at https://admin.fedoraproject.org/updates/xen-4.0.1-6.fc14 as it may still be possible to get it into the F14 release.
4.0.1-6.fc14 works for me!
-- Pasi
-- xen mailing list xen@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/xen
There is less urgency to test now because 4.0.1-6.fc14 has now made it into F14 in time for the release. You can of course continue to test if you want to.
Michael Young
Just wondering why kernel-2.6.32.23-170.xendom0.fc12.src.rpm cannot make itself into F14. It seems pretty stable. Why it cannot be just a package available via "yum install" . Anyway Fedora's Libvirt has the best compatibility with Xen 4.0.1. It's not too late switch back to Xen with PVOPS kernel. XenLinux 2.6.34.7 aka Suse is really dangerous . It outperforms pvops kernel about 15-20 % under Xen 4.0.1. I don't mention KVM performance. It's not a competitor for OpenSuse 11.3 after the most recent "zypper update"
Boris.
--- On Mon, 10/18/10, M A Young m.a.young@durham.ac.uk wrote:
From: M A Young m.a.young@durham.ac.uk Subject: Re: [Fedora-xen] another xen build To: xen@lists.fedoraproject.org Date: Monday, October 18, 2010, 6:17 AM
There is less urgency to test now because 4.0.1-6.fc14 has now made it into F14 in time for the release. You can of course continue to test if you want to.
Michael Young -- xen mailing list xen@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/xen
On Mon, Oct 18, 2010 at 07:49:29AM -0700, Boris Derzhavets wrote:
Just wondering why [1]kernel-2.6.32.23-170.xendom0.fc12.src.rpm cannot make itself into F14. It seems pretty stable. Why it cannot be just a package available via "yum install" . Anyway Fedora's Libvirt has the best compatibility with Xen 4.0.1. It's not too late switch back to Xen with PVOPS kernel. XenLinux 2.6.34.7 aka Suse is really dangerous . It outperforms pvops kernel about 15-20 % under Xen 4.0.1.
- How did you measure this performance difference? - Did you make sure kernel configuration is similar for both kernel types? Many pvops .configs have various debugging options enabled that kill the performance..
-- Pasi
I don't mention KVM performance. It's not a competitor for OpenSuse 11.3 after the most recent "zypper update"
Boris.
--- On Mon, 10/18/10, M A Young m.a.young@durham.ac.uk wrote:
From: M A Young <m.a.young@durham.ac.uk> Subject: Re: [Fedora-xen] another xen build To: xen@lists.fedoraproject.org Date: Monday, October 18, 2010, 6:17 AM There is less urgency to test now because 4.0.1-6.fc14 has now made it into F14 in time for the release. You can of course continue to test if you want to. Michael Young -- xen mailing list [2]xen@lists.fedoraproject.org [3]https://admin.fedoraproject.org/mailman/listinfo/xen
References
Visible links
- http://fedorapeople.org/%7Emyoung/dom0/src/kernel-2.6.32.23-170.xendom0.fc12...
- file:///mc/compose?to=xen@lists.fedoraproject.org
- https://admin.fedoraproject.org/mailman/listinfo/xen
-- xen mailing list xen@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/xen
- How did you measure this performance difference?
Database server running on PV DomU with intensive I/O. Workload connected with loading and unloading tables to/from ASCII files residing on the guest's image. > - Did you make sure kernel configuration is similar for both kernel types? > Many pvops .configs have various debugging options enabled that kill the performance..
It's my personal impression. I am not an expert in PVOPS config tuning, but try to avoid debugging options.
Boris.
--- On Wed, 10/20/10, Pasi Kärkkäinen pasik@iki.fi wrote:
From: Pasi Kärkkäinen pasik@iki.fi Subject: Re: [Fedora-xen] another xen build and pvops kernel To: "Boris Derzhavets" bderzhavets@yahoo.com Cc: xen@lists.fedoraproject.org, xen-users@lists.xensource.com Date: Wednesday, October 20, 2010, 1:42 PM
On Mon, Oct 18, 2010 at 07:49:29AM -0700, Boris Derzhavets wrote:
Just wondering why [1]kernel-2.6.32.23-170.xendom0.fc12.src.rpm cannot make itself into F14. It seems pretty stable. Why it cannot be just a package available via "yum install" . Anyway Fedora's Libvirt has the best compatibility with Xen 4.0.1. It's not too late switch back to Xen with PVOPS kernel. XenLinux 2.6.34.7 aka Suse is really dangerous . It outperforms pvops kernel about 15-20 % under Xen 4.0.1.
- How did you measure this performance difference? - Did you make sure kernel configuration is similar for both kernel types? Many pvops .configs have various debugging options enabled that kill the performance..
-- Pasi
I don't mention KVM performance. It's not a competitor for OpenSuse 11.3 after the most recent "zypper update"
Boris.
--- On Mon, 10/18/10, M A Young m.a.young@durham.ac.uk wrote:
From: M A Young m.a.young@durham.ac.uk Subject: Re: [Fedora-xen] another xen build To: xen@lists.fedoraproject.org Date: Monday, October 18, 2010, 6:17 AM
There is less urgency to test now because 4.0.1-6.fc14 has now made it into F14 in time for the release. You can of course continue to test if you want to.
Michael Young -- xen mailing list [2]xen@lists.fedoraproject.org [3]https://admin.fedoraproject.org/mailman/listinfo/xen
References
Visible links 1. http://fedorapeople.org/%7Emyoung/dom0/src/kernel-2.6.32.23-170.xendom0.fc12... 2. file:///mc/compose?to=xen@lists.fedoraproject.org 3. https://admin.fedoraproject.org/mailman/listinfo/xen
-- xen mailing list xen@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/xen
There is a new xen kernel (2.6.32.23-170.1.xendom0.fc12) available to download at http://koji.fedoraproject.org/koji/taskinfo?taskID=2547470 and at the repository http://repos.fedorapeople.org/repos/myoung/dom0-kernel/
It contains some patches added added to xen/stable-2.6.32.x since the last build, including an event channels fix.
Michael Young
On Fri, Oct 22, 2010 at 7:27 PM, M A Young m.a.young@durham.ac.uk wrote:
There is a new xen kernel (2.6.32.23-170.1.xendom0.fc12) available to download at http://koji.fedoraproject.org/koji/taskinfo?taskID=2547470 and at the repository http://repos.fedorapeople.org/repos/myoung/dom0-kernel/
It contains some patches added added to xen/stable-2.6.32.x since the last build, including an event channels fix.
Michael Young
take a look
https://bugzilla.redhat.com/show_bug.cgi?id=645252#c2
On Fri, 22 Oct 2010, Itamar Reis Peixoto wrote:
On Fri, Oct 22, 2010 at 7:27 PM, M A Young m.a.young@durham.ac.uk wrote:
There is a new xen kernel (2.6.32.23-170.1.xendom0.fc12) available to download at http://koji.fedoraproject.org/koji/taskinfo?taskID=2547470 and at the repository http://repos.fedorapeople.org/repos/myoung/dom0-kernel/
It contains some patches added added to xen/stable-2.6.32.x since the last build, including an event channels fix.
Michael Young
take a look
Try xen kernel-2.6.32.25-171.rc1.xendom0.fc12 instead at http://koji.fedoraproject.org/koji/taskinfo?taskID=2550132 or at the repository http://repos.fedorapeople.org/repos/myoung/dom0-kernel/ which has a fix for that and another security hole.
Michael Young
kernel-2.6.32.25-171.rc1.xendom0.fc12
RTC went mad. 1 sec = 1mS
Top was showing hours whizzing past in minutes. Windows VM went crazy doing stuff. FC13 VM went crazy too doing all sorts of things really quickly.
Bizzare effect.
Went back to -168 kernel. Alls well again.
Cheers V96
On Sun, 24 Oct 2010 03:50:40 am M A Young wrote:
On Fri, 22 Oct 2010, Itamar Reis Peixoto wrote:
On Fri, Oct 22, 2010 at 7:27 PM, M A Young m.a.young@durham.ac.uk wrote:
There is a new xen kernel (2.6.32.23-170.1.xendom0.fc12) available to download at http://koji.fedoraproject.org/koji/taskinfo?taskID=2547470 and at the repository http://repos.fedorapeople.org/repos/myoung/dom0-kernel/
It contains some patches added added to xen/stable-2.6.32.x since the last build, including an event channels fix.
Michael Young
take a look
Try xen kernel-2.6.32.25-171.rc1.xendom0.fc12 instead at http://koji.fedoraproject.org/koji/taskinfo?taskID=2550132 or at the repository http://repos.fedorapeople.org/repos/myoung/dom0-kernel/ which has a fix for that and another security hole.
Michael Young
On Sat, Oct 23, 2010 at 11:50 AM, M A Young m.a.young@durham.ac.uk wrote:
On Fri, 22 Oct 2010, Itamar Reis Peixoto wrote:
On Fri, Oct 22, 2010 at 7:27 PM, M A Young m.a.young@durham.ac.uk wrote:
There is a new xen kernel (2.6.32.23-170.1.xendom0.fc12) available to download at http://koji.fedoraproject.org/koji/taskinfo?taskID=2547470 and at the repository http://repos.fedorapeople.org/repos/myoung/dom0-kernel/
It contains some patches added added to xen/stable-2.6.32.x since the last build, including an event channels fix.
Michael Young
take a look
Try xen kernel-2.6.32.25-171.rc1.xendom0.fc12 instead at http://koji.fedoraproject.org/koji/taskinfo?taskID=2550132 or at the repository http://repos.fedorapeople.org/repos/myoung/dom0-kernel/ which has a fix for that and another security hole.
Anyone else seeing problems with software raid? On a new Fedora 14 system, the distro kernel is fine, but my 4 SCSI drive system has md devices fail with xendom0...
Oct 27 23:31:17 elm kernel: [ 40.504212] dracut: mdadm: /dev/md0 has been started with 2 drives. Oct 27 23:31:17 elm kernel: [ 40.590546] dracut: mdadm: /dev/md1 has been started with 2 drives. Oct 27 23:31:17 elm kernel: [ 40.629581] dracut: mdadm: /dev/md2 has been started with 3 drives. Oct 27 23:31:17 elm kernel: [ 40.709149] dracut: mdadm: /dev/md3 has been started with 3 drives. [snip] Oct 27 23:39:55 elm kernel: dracut: mdadm: /dev/md0 has been started with 2 drives. Oct 27 23:39:55 elm kernel: dracut: mdadm: /dev/md1 has been started with 2 drives. Oct 27 23:39:55 elm kernel: mdadm[617]: segfault at 46 ip 006b7997 sp bfe4b2d0 error 4 in libc-2.12.90.so[658000+18d000] Oct 27 23:39:55 elm kernel: mdadm[636]: segfault at 46 ip 006b7997 sp bfa19490 error 4 in libc-2.12.90.so[658000+18d000]
The "Oct 27 23:31:17" messages reflect fc14, while "Oct 27 23:39:55" is xendom0. The system boots fine, but just won't see the extra meta devices. Maybe just can't be done with a .32 kernel on fc14?
jerry
On Thu, Oct 28, 2010 at 1:03 AM, Jerry Amundson jamundso@gmail.com wrote:
On Sat, Oct 23, 2010 at 11:50 AM, M A Young m.a.young@durham.ac.uk wrote:
Try xen kernel-2.6.32.25-171.rc1.xendom0.fc12 instead at http://koji.fedoraproject.org/koji/taskinfo?taskID=2550132 or at the repository http://repos.fedorapeople.org/repos/myoung/dom0-kernel/ which has a fix for that and another security hole.
Anyone else seeing problems with software raid? On a new Fedora 14 system, the distro kernel is fine, but my 4 SCSI drive system has md devices fail with xendom0...
Oct 27 23:31:17 elm kernel: [ 40.504212] dracut: mdadm: /dev/md0 has been started with 2 drives. Oct 27 23:31:17 elm kernel: [ 40.590546] dracut: mdadm: /dev/md1 has been started with 2 drives. Oct 27 23:31:17 elm kernel: [ 40.629581] dracut: mdadm: /dev/md2 has been started with 3 drives. Oct 27 23:31:17 elm kernel: [ 40.709149] dracut: mdadm: /dev/md3 has been started with 3 drives. [snip] Oct 27 23:39:55 elm kernel: dracut: mdadm: /dev/md0 has been started with 2 drives. Oct 27 23:39:55 elm kernel: dracut: mdadm: /dev/md1 has been started with 2 drives. Oct 27 23:39:55 elm kernel: mdadm[617]: segfault at 46 ip 006b7997 sp bfe4b2d0 error 4 in libc-2.12.90.so[658000+18d000] Oct 27 23:39:55 elm kernel: mdadm[636]: segfault at 46 ip 006b7997 sp bfa19490 error 4 in libc-2.12.90.so[658000+18d000]
The "Oct 27 23:31:17" messages reflect fc14, while "Oct 27 23:39:55" is xendom0. The system boots fine, but just won't see the extra meta devices.
I was tripped up by my own kernel hopping - the current mdadm.conf wasn't present in the initramfs made with the kernel install done earlier. A mkinitrd for the needful kernel, and poof! the raid devices are fine.
jerry
I am having trouble with F14, and this and the previous kernel:
The problem seems to start around line 687: xen_balloon: Initialising balloon driver with page order 0. last_pfn = 0x444690 max_arch_pfn = 0x400000000 BUG: unable to handle kernel paging request at ffffea000e700030 IP: [<ffffffff812351b7>] __list_add+0x6c/0x81
but I am not sure.
Here is the really weird thing, I have this system: http://www.supermicro.com/Aplus/system/2U/2022/AS-2022TG-HTRF.cfm with 4 machines using these mbs (with 1 AMD 6172 each): http://www.supermicro.com/Aplus/motherboard/Opteron6100/SR56x0/H8DGT-HF.cfm lshw output here: http://pastie.org/1242726 and one of the 4 systems is able to boot the system using xen but the others all die at the same place (using the same disks to test all 4). I have tried changing as many BIOS settings as I can - all the ACPI ones and many others and I can get any change. (BIOS sucks btw, you can't print or output the full config in any useful way to diff/track/compare them - argh!)
Here: http://pastie.org/1242727 is the output of the one that does work, unfortunately as I monkeyed around with it I think it is now not finding the igb NICs :(
The systems come up fine with both the current F14 kernel and both the latest and previous dom0 kernel without xen # uname -a Linux xentest 2.6.32.23-170.1.xendom0.fc12.x86_64 #1 SMP Thu Oct 21 19:32:14 UTC 2010 x86_64 x86_64 x86_64 GNU/Linux
Name : xen Arch : x86_64 Version : 4.0.1 Release : 6.fc14
I tried a number of grub configs some with various debug and acpi/apic settings but no luck, here is a simple one when I was trying the new kernel:
title Xen root (hd1,0) kernel /xen.gz dom0_mem=1024M loglvl=all guest_loglvl=all module /vmlinuz-2.6.32.23-170.1.xendom0.fc12.x86_64 ro root=UUID=84e6a109-84c5-4288-80de-ea368a224a3a rd_MD_UUID=c557e3e8:d9ef50b8:b5e9861f:26975533 rd_MD_UUID=5825fd8c:77ea8324:ae0c8804:8a0c248 7 rd_NO_LUKS rd_NO_LVM rd_NO_DM LANG=en_US.UTF-8 SYSFONT=latarcyrheb-sun16 KEYTABLE=us nomodeset initrd /initramfs-2.6.32.23-170.1.xendom0.fc12.x86_64.img
Maybe I missed a step somewhere but with the xen 4.0.1.x in F14 and the dom0 kernel repo, I thought that there was no more manual compiling, etc - maybe I was mistaken.
Any help/idea would be greatly appreciated. Thanks.
On 10/22/10 2:27 PM, M A Young wrote:
There is a new xen kernel (2.6.32.23-170.1.xendom0.fc12) available to download at http://koji.fedoraproject.org/koji/taskinfo?taskID=2547470 and at the repository http://repos.fedorapeople.org/repos/myoung/dom0-kernel/
It contains some patches added added to xen/stable-2.6.32.x since the last build, including an event channels fix.
Michael Young
xen mailing list xen@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/xen
On Sat, 23 Oct 2010, fcxen user wrote:
I am having trouble with F14, and this and the previous kernel:
The problem seems to start around line 687: xen_balloon: Initialising balloon driver with page order 0. last_pfn = 0x444690 max_arch_pfn = 0x400000000 BUG: unable to handle kernel paging request at ffffea000e700030 IP: [<ffffffff812351b7>] __list_add+0x6c/0x81
Did earlier kernels work or not or didn't you try them? Also do you have a log from the 170.1 kernel or the newly built 171 kernel? These still have the debuginfo packages available which means I can get a bit more information about the crash.
Michael Young
170, 170.1, 171 crashed the same
using: title Xen root (hd0,0) kernel /xen.gz loglvl=all guest_loglvl=all sync_console console_to_ring com2=115200,8n1 console=com2,vga lapic=debug apic_verbosity=debug apic=debug iommu=off module /vmlinuz-2.6.32.25-171.rc1.xendom0.fc12.x86_64 ro root=UUID=84e6a109-84c5-4288-80de-ea368a224a3a rd_MD_UUID=c557e3e8:d9ef50b8:b5e9861f:26975533 rd_MD_UUID=5825fd8c:77ea8324:ae0c8804:8a0c2 487 rd_NO_LUKS rd_NO_LVM rd_NO_DM LANG=en_US.UTF-8 SYSFONT=latarcyrheb-sun16 KEYTABLE=us console=hvc0 earlyprintk=xen nomodeset initcall_debug debug loglevel=10 module /initramfs-2.6.32.25-171.rc1.xendom0.fc12.x86_64.img
Latest output is: http://pastie.org/1243649
I could give you a login on the box if that would help...
Let me know if I should try other debug options/settings, thanks.
On 10/23/10 10:04 AM, M A Young wrote:
On Sat, 23 Oct 2010, fcxen user wrote:
I am having trouble with F14, and this and the previous kernel:
The problem seems to start around line 687: xen_balloon: Initialising balloon driver with page order 0. last_pfn = 0x444690 max_arch_pfn = 0x400000000 BUG: unable to handle kernel paging request at ffffea000e700030 IP: [<ffffffff812351b7>] __list_add+0x6c/0x81
Did earlier kernels work or not or didn't you try them? Also do you have a log from the 170.1 kernel or the newly built 171 kernel? These still have the debuginfo packages available which means I can get a bit more information about the crash.
Michael Young
xen@lists.stg.fedoraproject.org