- openSUSE Virtual

[opensuse-virtual] status of kernel-xen in Kernel:stable?
by PGNet Dev 21 Nov '15

21 Nov '15

I note that as of today, 'kernel-xen' is no longer included in builds @ https://build.opensuse.org/project/show/Kernel:stable Is its removal an oversight, or planned? IIRC, there were plans for the migration to kernel-pv (also missing, atm) -- and, eventually, to a single, consolidated, pv-ready kernel. Has that occurrred, perhaps? -- To unsubscribe, e-mail: opensuse-virtual+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-virtual+owner(a)opensuse.org

1 1

[opensuse-virtual] Chainloaded Xen Dom0 on EFI hardware FAILs at grub config generation after upgrade from Opensuse 13.2 -> Leap 42.1. Has the Xen-on-EFI prep process changed?
by PGNet Dev 20 Nov '15

20 Nov '15

I've been running a Xen Dom0 server Opensuse 13.2 Xen from /Virtualization kernel from Kernel:Stable rpm -qa | egrep "kernel-xen-4|^xen-4|grub2-x" | sort grub2-x86_64-efi-2.02~beta2-68.2.x86_64 grub2-x86_64-xen-2.02~beta2-68.2.x86_64 kernel-xen-4.3.0-4.1.g734b32c.x86_64 xen-4.6.0_02-397.1.x86_64 on EFI hardware. I've been chainloading xen.efi. I'm attempting an upgrade on the server, from Opensuse 13.2 -> Leap. After the upgrade, but before a reboot, generation of the grub config now FAILs during the mkinitrd step(s) @ console /usr/sbin/grub2-mkconfig Generating grub configuration file ... # # DO NOT EDIT THIS FILE # # It is automatically generated by grub2-mkconfig using templates # from /etc/grub.d and settings from /etc/default/grub # ### BEGIN /etc/grub.d/00_header ### if [ -f ${config_directory}/grubenv ]; then load_env -f ${config_directory}/grubenv elif [ -s $prefix/grubenv ]; then load_env fi if [ "${env_block}" ] ; then load_env -f "${env_block}" fi if [ "${next_entry}" ] ; then set default="${next_entry}" set next_entry= save_env next_entry if [ "${env_block}" ] ; then save_env -f "${env_block}" next_entry fi set boot_once=true else set default="${saved_entry}" fi if [ x"${feature_menuentry_id}" = xy ]; then menuentry_id_option="--id" else menuentry_id_option="" fi export menuentry_id_option if [ "${prev_saved_entry}" ]; then set saved_entry="${prev_saved_entry}" save_env saved_entry set prev_saved_entry= save_env prev_saved_entry set boot_once=true fi function savedefault { if [ -z "${boot_once}" ]; then saved_entry="${chosen}" save_env saved_entry fi } function load_video { if [ x$feature_all_video_module = xy ]; then insmod all_video else insmod efi_gop insmod efi_uga insmod ieee1275_fb insmod vbe insmod vga insmod video_bochs insmod video_cirrus fi } serial --unit=0 --speed=115200 --word=8 --parity=no --stop=1 terminal_input console serial terminal_output console serial if [ x${boot_once} = xtrue ]; then set timeout=0 elif [ x$feature_timeout_style = xy ] ; then set timeout_style=menu set timeout=8 # Fallback normal timeout code in case the timeout_style feature is # unavailable. else set timeout=8 fi ### END /etc/grub.d/00_header ### ... ### BEGIN /etc/grub.d/20_linux_xen ### Found hypervisor: /boot/xen-4.6.0_02-397.gz Found linux image: /boot/vmlinuz-4.3.0-4.g734b32c-xen Found initrd image: /boot/initrd-4.3.0-4.g734b32c-xen menuentry 'openSUSE Leap 42.1, with Xen hypervisor' --class opensuse --class gnu-linux --class gnu --class os --class xen $menuentry_id_option 'xen-gnulinux-simple-4716053b-39da-4cc1-a7ce-72171b862357' { echo 'Loading Xen 4.6.0_02-397 with Linux 4.3.0-4.g734b32c-xen ...' chainloader $cmdpath/xen-4.6.0_02-397.efi xen-4.6.0_02-397.efi config.1 } cp: cannot stat '/vmlinuz-4.3.0-4.g734b32c-xen': No such file or directory Where ls -al /boot/efi/EFI/opensuse total 2.3M drwxrwxr-x 2 root root 4.0K Nov 16 17:16 ./ drwxrwxr-x 4 root root 4.0K Mar 28 2015 ../ -rwxrwxr-x 1 root root 133K Nov 16 15:26 grubx64.efi* -rwxrwxr-x 1 root root 42 Nov 16 17:16 grub.xen-files* -rwxrwxr-x 1 root root 990 Nov 16 17:16 xen-4.6.0_02-397.cfg* -rwxrwxr-x 1 root root 2.1M Nov 11 13:18 xen-4.6.0_02-397.efi* Have the prep steps needed changed? -- To unsubscribe, e-mail: opensuse-virtual+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-virtual+owner(a)opensuse.org

2 11

[opensuse-virtual] 'Virtualization' repo for Leap -- name, and activate?
by PGNet Dev 06 Nov '15

06 Nov '15

I currently run/test xen-4.5.1_13-391.1.x86_64 <-- d.o.o://Virtualization/openSUSE_13.2 kernel-xen 4.3.0-1.g7b374a4-xen <-- d.o.o:///Kernel:/stable/standard on Opensuse 13.2. With Leap release, I see https://build.opensuse.org/project/repository_state/Virtualization/openSUSE… , currently unpopulated. For other Leap repos being activated, the naming convention appears to be: .../openSUSE_Leap_42.1 (1) Should the Virtualization repo be similarly named? (2) Can whatever the correctly-named repo is to be get activated/populated? , if not already in the works ... -- To unsubscribe, e-mail: opensuse-virtual+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-virtual+owner(a)opensuse.org

3 7

[opensuse-virtual] pvgrub2-boot of Opensuse Xen DomU -- possible? documentation? example?
by PGNet Dev 05 Nov '15

05 Nov '15

I'd recently attempted to boot an Opensuse DomU, on Opensuse Dom0, switching to pvgrub2 rather than pygrub as the bootloader mechanism. Failed at it, and asked at the Xen ML "pvgrub2 launch of Xen 4.5.1 DomU just drops to grub2 prompt ?" http://xen.markmail.org/thread/dmugrhbg6rv4vzn7 Before proceeding to hunt-n-peck my way through figuring this out, Is DomU on current Opensuse+Xen, in my case, rpm -qa | egrep "kernel-xen-4|^xen-4|grub2-x" | sort grub2-x86_64-efi-2.02~beta2-20.10.1.x86_64 grub2-x86_64-xen-2.02~beta2-20.10.1.x86_64 kernel-xen-4.2.5-2.1.g0491388.x86_64 xen-4.5.1_13-391.1.x86_64 pvgrub2-bootable? If YES, is there any clear *documentation*, ideally with an example? -- To unsubscribe, e-mail: opensuse-virtual+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-virtual+owner(a)opensuse.org

4 13

[opensuse-virtual] status/plans for direct0boot (not chainload) of Xen Dom0 on UEFI hardware?
by PGNet Dev 30 Oct '15

30 Oct '15

I run an Opensuse 13.2 Xen server. Currently, the following are installed & active in Dom0. rpm -qa | grep -i xen | sort grub2-x86_64-xen-2.02~beta2-20.10.1.x86_64 kernel-xen-4.2.5-1.1.g27d2719.x86_64 xen-4.5.1_10-390.1.x86_64 xen-libs-4.5.1_10-390.1.x86_64 xen-tools-4.5.1_10-390.1.x86_64 The server mobo is UEFI capable. It's configured to boot Dom0 from UEFI via still-necessary chainload tree /boot/efi/ /boot/efi/ └── EFI ├── opensuse │ └── grubx64.efi └── XEN ├── initrd-xen ├── vmlinuz-xen ├── xen.cfg └── xen.efi etc Works fine. IIUC, (1) Xen 4.6 is arriving soon (2) Kernel 4.3+ is arriving soon (3) Opensuse's moving from patched kernel-xen to pvops-integrated kernel for Dom0/DomU Will any of that^^ progress result in DIRECT-booting of Opensuse Dom0 on EUFI hardware, removing the necessity to chainload? -- To unsubscribe, e-mail: opensuse-virtual+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-virtual+owner(a)opensuse.org

2 5

[opensuse-virtual] Xen Critical vulnerability CVE-2015-7835 unpatched in Opensuse/Xen packages?
by PGNet Dev 29 Oct '15

29 Oct '15

I run latest Xen from d.o.o's Virtualization/openSUSE_13.2 repo rpm -qa | grep -i ^xen | sort xen-4.5.1_10-390.1.x86_64 xen-libs-4.5.1_10-390.1.x86_64 xen-tools-4.5.1_10-390.1.x86_64 Xen's now made public it's latest critical advisory http://arstechnica.com/security/2015/10/xen-patches-7-year-old-bug-that-sha… "Xen patches 7-year-old bug that shattered hypervisor security. Critical vulnerability allowed some guests to access underlying operating system." http://xenbits.xen.org/xsa/advisory-148.html Advisory XSA-148 Public release 2015-10-29 11:59 ... CVE(s) CVE-2015-7835 Title x86: Uncontrolled creation of large page mappings by PV guests The advisory instructs patching to resolve RESOLUTION ========== Applying the appropriate attached patch resolves this issue. xsa148.patch xen-unstable, Xen 4.6.x xsa148-4.5.patch Xen 4.5.x xsa148-4.4.patch Xen 4.4.x, Xen 4.3.x Checking installed Xen's changelog rpm -q --changelog xen | egrep "CVE-2015-7835|xsa148" (empty) it's not been applied. Or, afaict from obs, even submitted. Where's this security patch in the package tree? -- To unsubscribe, e-mail: opensuse-virtual+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-virtual+owner(a)opensuse.org

1 0

[opensuse-virtual] after changes to 'xen.cfg' on a grub2 chainloaded Xen instance, what addl steps (mkconfig? mkinitrd? cp?) need to be exec'd?
by lyndat3＠your-mail.com 23 Apr '15

23 Apr '15

I boot Xen on UEFI + grub2. uname -rm 3.19.4-1.g74c332b-xen x86_64 rpm -qa | grep -i xen-4 xen-4.5.0_03-367.1.x86_64 Xen is chainloaded from grub2 config cat /boot/grub2/custom.cfg menuentry 'xen' { ... chainloader /efi/xen/xen.efi } cat /boot/efi/efi/xen/xen.cfg [global] default=xen [xen] options=... kernel=vmlinuz-xen ... ramdisk=initrd-xen ... each time a change is made to 'xen.cfg' which of these steps need to be executed to propagate fully to next boot? grub2-mkconfig -o /boot/grub2/grub.cfg mkinitrd /bin/cp -f /usr/lib64/efi/xen.efi /boot/vmlinuz-xen /boot/initrd-xen /boot/efi/efi/xen/ all of them? in that order? LT -- To unsubscribe, e-mail: opensuse-virtual+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-virtual+owner(a)opensuse.org

3 3

[opensuse-virtual] regular updates to Xen pkgs, but no changes in 'changelog'. Where are differences viewable?
by lyndat3＠your-mail.com 21 Apr '15

21 Apr '15

I'm seeing regular upgrades of Xen pkgs from http://download.opensuse.org/repositories/Virtualization/openSUSE_13.2 For example, from zypper history 2015-04-14 17:52:43|install|xen|4.5.0_03-366.5|x86_64||XEN4|f41fd5e7cf75ce69ba159ef1282f5905af96acea9df3f256d11afb072236e562| then, today 2015-04-20 17:32:03|install|xen|4.5.0_03-366.6|x86_64||XEN4|d2bd0baa5bbe27560bfb8e2bf32e1581199fe8b3e01404a556dd465677d56b73| Checking the changelogs rpm -q --changelog xen after each upgrade, I'm not seeing any difference. This happens often. Where are the actual differences logged? -- To unsubscribe, e-mail: opensuse-virtual+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-virtual+owner(a)opensuse.org

3 2

[opensuse-virtual] configuring the chainloader 'workaround' for grub2 booting of Xen on EFI ?
by lyndat3＠your-mail.com 13 Apr '15

13 Apr '15

I'm setting up GRUB2-booting of Xen on EFI Reading http://lists.xenproject.org/archives/html/xen-users/2014-03/msg00142.html http://wiki.xenproject.org/wiki/Xen_EFI Booting Xen.gz via GRUB2 can't be done directly, but needs a chainloader workaround. The example GRUB2 config given is: menuentry 'Xen EFI' { insmod part_gpt insmod search_fs_uuid insmod chain chainloader (hd0,gpt1)/EFI/XEN/xen.efi } On EFI opensuse with grub2 & Xen installed, find / | egrep "xen|grub" | grep "\.efi$" /boot/grub2/x86_64-efi/grub.efi /boot/grub2/x86_64-efi/core.efi /boot/efi/EFI/opensuse/grubx64.efi /usr/lib/grub2/x86_64-efi/grub.efi /usr/lib64/efi/grub.efi /usr/lib64/efi/xen.efi /usr/lib64/efi/xen-4.efi /usr/lib64/efi/xen-4.5.0_03-359.efi /usr/lib64/efi/xen-4.5.efi The xen.efi file is located /usr/lib64/efi/xen.efi not at the target path chainloader (hd0,gpt1)/EFI/XEN/xen.efi (1) Is the right procedure to simply mkdir -p /boot/efi/EFI/XEN cp /usr/lib64/efi/xen.efi /boot/efi/EFI/XEN/ (2) If the Xen install doesn't populate xen.efi into that directory, then at every Xen update/upgrade, will xen.efi need to be re-copied manually? (3) With the GRUB2-installed opensuse, I find no more 'device.map' file under /boot/... Is device.map still used? If none exists, what's the default mapping? For a given location, how do you query/return the device-map drive ? E.g., grub2-probe --target=device /boot/efi returns /dev/sdc2 In order to correctly modify the example's "(hd0,gpt1)" in chainloader (hd0,gpt1)/EFI/XEN/xen.efi I need to know whether /dev/sdc2 is (hd0) or (hd1) etc. LT -- To unsubscribe, e-mail: opensuse-virtual+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-virtual+owner(a)opensuse.org

2 4

[opensuse-virtual] Intermittent crash on reboot of opensuse 13.2 + Xen 4.5 ?
by lyndat3＠your-mail.com 13 Apr '15

13 Apr '15

I can't reproduce this on-demand. It doesn't happen every time. But rebooting a running Xen instance does crash - occasionally - and when it does, it reports this in serial console ... (XEN) [2015-04-08 16:15:05] irq.c:2120: dom0: forcing unbind of pirq 201 (XEN) [2015-04-08 16:15:05] irq.c:2120: dom0: forcing unbind of pirq 202 (XEN) [2015-04-08 16:15:05] irq.c:2120: dom0: forcing unbind of pirq 203 (XEN) [2015-04-08 16:15:05] irq.c:2120: dom0: forcing unbind of pirq 190 [38366.449320] e1000e: EEE TX LPI TIMER: 00000011 [38366.468732] reboot: Restarting system (XEN) [2015-04-08 16:15:05] Domain 0 shutdown: rebooting machine. (XEN) [2015-04-08 16:15:06] ----[ Xen-4.5.0_03-363 x86_64 debug=n Not tainted ]---- (XEN) [2015-04-08 16:15:06] CPU: 0 (XEN) [2015-04-08 16:15:06] RIP: e008:[<000000009e6c4000>] 000000009e6c4000 (XEN) [2015-04-08 16:15:06] RFLAGS: 0000000000010247 CONTEXT: hypervisor (XEN) [2015-04-08 16:15:06] rax: 000000009e670340 rbx: 0000000000000000 rcx: 0000000000000000 (XEN) [2015-04-08 16:15:06] rdx: 0000000000000000 rsi: 0000000000000000 rdi: 0000000000000000 (XEN) [2015-04-08 16:15:06] rbp: 0000000000000000 rsp: ffff82d080457dd0 r8: 0000000000000000 (XEN) [2015-04-08 16:15:06] r9: 0000000000000000 r10: 0000000000000000 r11: 0000000000000008 (XEN) [2015-04-08 16:15:06] r12: 0000000000000000 r13: 0000000000000061 r14: 00000000fee1dead (XEN) [2015-04-08 16:15:06] r15: 0000000000000000 cr0: 0000000080050033 cr4: 00000000001526f0 (XEN) [2015-04-08 16:15:06] cr3: 00000008459d5000 cr2: 000000009e6c4000 (XEN) [2015-04-08 16:15:06] ds: 0000 es: 0000 fs: 0000 gs: 0000 ss: e010 cs: e008 (XEN) [2015-04-08 16:15:06] Xen stack trace from rsp=ffff82d080457dd0: (XEN) [2015-04-08 16:15:06] 000000009efe42f6 00000000fee1dead ffff82d0802278a4 efff00000000000a (XEN) [2015-04-08 16:15:06] ffff82d080262000 00000008459d5000 ffff82d080227a7a 000000069ed2a000 (XEN) [2015-04-08 16:15:06] 0000000000000000 0000000000152670 0000000000000700 0000000000000061 (XEN) [2015-04-08 16:15:06] 0000000000000000 00000000fffffffe ffff82d08018477c ffff82d080457e98 (XEN) [2015-04-08 16:15:06] 0000000080457e58 000036363338335b 0000000000000000 0000000000000001 (XEN) [2015-04-08 16:15:06] 0000000000000001 ffff830845989000 ffff830845989138 00000000fee1dead (XEN) [2015-04-08 16:15:06] ffff82d080129c99 0000000000a0fb00 ffff82d080105721 ffffffffffffffff (XEN) [2015-04-08 16:15:06] 0000000000000000 0000000028121969 ffffffff80a1f100 0000000000000002 (XEN) [2015-04-08 16:15:06] ffff82d080128d3f 000000010007c000 ffff83009e786000 0000000000000029 (XEN) [2015-04-08 16:15:06] ffffffff80ab47e0 ffff8800bea23b1c ffff83009e786000 0000000028121969 (XEN) [2015-04-08 16:15:06] ffff82d080224119 0000000000000000 0000000000000147 0000000000000008 (XEN) [2015-04-08 16:15:06] 0000000000000000 0000000028121969 0000000000000000 0000000000000282 (XEN) [2015-04-08 16:15:06] 0000000000000064 0000000000000523 0000000000000000 000000000000001d (XEN) [2015-04-08 16:15:06] ffffffff800113aa 0000000000000001 ffff8800a9f93e1c 0000000000000002 (XEN) [2015-04-08 16:15:06] 0001010000000000 ffffffff800113aa 000000000000e033 0000000000000282 (XEN) [2015-04-08 16:15:06] ffff8800a9f93df8 000000000000e02b 0000000000000000 0000000000000000 (XEN) [2015-04-08 16:15:06] 0000000000000000 0000000000000000 0000000000000000 ffff83009e786000 (XEN) [2015-04-08 16:15:06] 0000000000000000 0000000000000000 (XEN) [2015-04-08 16:15:06] Xen call trace: (XEN) [2015-04-08 16:15:06] [<000000009e6c4000>] 000000009e6c4000 (XEN) [2015-04-08 16:15:06] [<ffff82d0802278a4>] efi_rs_enter+0xf4/0x110 (XEN) [2015-04-08 16:15:06] [<ffff82d080227a7a>] efi_reset_system+0x3a/0x60 (XEN) [2015-04-08 16:15:06] [<ffff82d08018477c>] machine_restart+0xcc/0x220 (XEN) [2015-04-08 16:15:06] [<ffff82d080129c99>] hwdom_shutdown+0x89/0x90 (XEN) [2015-04-08 16:15:06] [<ffff82d080105721>] domain_shutdown+0xf1/0x100 (XEN) [2015-04-08 16:15:06] [<ffff82d080128d3f>] do_sched_op+0x1af/0x440 (XEN) [2015-04-08 16:15:06] [<ffff82d080224119>] syscall_enter+0xa9/0xae (XEN) [2015-04-08 16:15:06] (XEN) [2015-04-08 16:15:06] Pagetable walk from 000000009e6c4000: (XEN) [2015-04-08 16:15:06] L4[0x000] = 00000008459d4063 ffffffffffffffff (XEN) [2015-04-08 16:15:07] L3[0x002] = 000000008c674063 ffffffffffffffff (XEN) [2015-04-08 16:15:07] L2[0x0f3] = 000000009e5ff063 ffffffffffffffff (XEN) [2015-04-08 16:15:07] L1[0x0c4] = 0000000000000000 ffffffffffffffff (XEN) [2015-04-08 16:15:07] (XEN) [2015-04-08 16:15:07] **************************************** (XEN) [2015-04-08 16:15:07] Panic on CPU 0: (XEN) [2015-04-08 16:15:07] FATAL PAGE FAULT (XEN) [2015-04-08 16:15:07] [error_code=0010] (XEN) [2015-04-08 16:15:07] Faulting linear address: 000000009e6c4000 (XEN) [2015-04-08 16:15:07] **************************************** (XEN) [2015-04-08 16:15:07] (XEN) [2015-04-08 16:15:07] Manual reset required ('noreboot' specified) Is this a known or obvious problem? I'll try to figure out how to reliably reproduc it. LT -- To unsubscribe, e-mail: opensuse-virtual+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse-virtual+owner(a)opensuse.org

3 5