Hallo, I have updated system to xen 4.2.1, linux kernel 3.7.5. System boots without problem. When I run reboot or shutdown -r now, system correctly shutdowns all runlevels and then system crashed, see crash_output.txt. I have the same problem with kernel linux 3.2.37. When I shudown system (halt), then it is without problem. I have tested kernel 3.7.5 without xen hypervisor, it was without problems. Required debug informations (http://wiki.xen.org/wiki/XenParavirtOpsHelp) in attachment. Could you please help me with this problem? Thanks Michal _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
>>> On 26.02.13 at 15:32, Michal Fiala <fiala@mfiala.net> wrote: > I have updated system to xen 4.2.1, linux kernel 3.7.5. System boots > without problem. When I run reboot or shutdown -r now, system correctly > shutdowns all runlevels and then system crashed, see crash_output.txt. > I have the same problem with kernel linux 3.2.37. When I shudown system > (halt), then it is without problem. I have tested kernel 3.7.5 without > xen hypervisor, it was without problems. Required debug informations > (http://wiki.xen.org/wiki/XenParavirtOpsHelp) in attachment. > Could you please help me with this problem?It''s apparently going the triple fault path, and load_idt() under Xen doesn''t do what the code here expects (i.e. that reboot method just can''t work under Xen). Are you overriding the reboot type on the kernel command line? Jan
The kernel command line from grub.conf kernel /xen.gz dom0_mem=4G,max:4G dom0_max_vcpus=4 dom0_vcpus_pin console=com2,vga com2=115200,8n1 module /linux-3.7.5-hardened root=/dev/sda6 console=tty0 console=hvc0 I have not modified reboot type. On 02/26/2013 05:05 PM, Jan Beulich wrote:>>>> On 26.02.13 at 15:32, Michal Fiala <fiala@mfiala.net> wrote: >> I have updated system to xen 4.2.1, linux kernel 3.7.5. System boots >> without problem. When I run reboot or shutdown -r now, system correctly >> shutdowns all runlevels and then system crashed, see crash_output.txt. >> I have the same problem with kernel linux 3.2.37. When I shudown system >> (halt), then it is without problem. I have tested kernel 3.7.5 without >> xen hypervisor, it was without problems. Required debug informations >> (http://wiki.xen.org/wiki/XenParavirtOpsHelp) in attachment. >> Could you please help me with this problem? > > It''s apparently going the triple fault path, and load_idt() under Xen > doesn''t do what the code here expects (i.e. that reboot method > just can''t work under Xen). Are you overriding the reboot type on > the kernel command line? > > Jan >
>>> On 26.02.13 at 17:28, Michal Fiala <fiala@mfiala.net> wrote:Please don''t top post.> The kernel command line from grub.conf > > kernel /xen.gz dom0_mem=4G,max:4G dom0_max_vcpus=4 dom0_vcpus_pin > console=com2,vga com2=115200,8n1 > module /linux-3.7.5-hardened root=/dev/sda6 console=tty0 console=hvc0 > > I have not modified reboot type.And did you check your full boot log for an info-level message like "%s series board detected. Selecting %s-method for reboot.\n" Jan> On 02/26/2013 05:05 PM, Jan Beulich wrote: >>>>> On 26.02.13 at 15:32, Michal Fiala <fiala@mfiala.net> wrote: >>> I have updated system to xen 4.2.1, linux kernel 3.7.5. System boots >>> without problem. When I run reboot or shutdown -r now, system correctly >>> shutdowns all runlevels and then system crashed, see crash_output.txt. >>> I have the same problem with kernel linux 3.2.37. When I shudown system >>> (halt), then it is without problem. I have tested kernel 3.7.5 without >>> xen hypervisor, it was without problems. Required debug informations >>> (http://wiki.xen.org/wiki/XenParavirtOpsHelp) in attachment. >>> Could you please help me with this problem? >> >> It''s apparently going the triple fault path, and load_idt() under Xen >> doesn''t do what the code here expects (i.e. that reboot method >> just can''t work under Xen). Are you overriding the reboot type on >> the kernel command line? >> >> Jan >>
On 02/26/2013 05:33 PM, Jan Beulich wrote:>>>> On 26.02.13 at 17:28, Michal Fiala <fiala@mfiala.net> wrote: > > Please don''t top post. > >> The kernel command line from grub.conf >> >> kernel /xen.gz dom0_mem=4G,max:4G dom0_max_vcpus=4 dom0_vcpus_pin >> console=com2,vga com2=115200,8n1 >> module /linux-3.7.5-hardened root=/dev/sda6 console=tty0 console=hvc0 >> >> I have not modified reboot type. > > And did you check your full boot log for an info-level message like > > "%s series board detected. Selecting %s-method for reboot.\n"I have only checked normal boot log, see attachment in my first mail. How can I enable info-level message?> > Jan > >> On 02/26/2013 05:05 PM, Jan Beulich wrote: >>>>>> On 26.02.13 at 15:32, Michal Fiala <fiala@mfiala.net> wrote: >>>> I have updated system to xen 4.2.1, linux kernel 3.7.5. System boots >>>> without problem. When I run reboot or shutdown -r now, system correctly >>>> shutdowns all runlevels and then system crashed, see crash_output.txt. >>>> I have the same problem with kernel linux 3.2.37. When I shudown system >>>> (halt), then it is without problem. I have tested kernel 3.7.5 without >>>> xen hypervisor, it was without problems. Required debug informations >>>> (http://wiki.xen.org/wiki/XenParavirtOpsHelp) in attachment. >>>> Could you please help me with this problem? >>> >>> It''s apparently going the triple fault path, and load_idt() under Xen >>> doesn''t do what the code here expects (i.e. that reboot method >>> just can''t work under Xen). Are you overriding the reboot type on >>> the kernel command line? >>> >>> Jan >>> > > > > > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xen.org > http://lists.xen.org/xen-devel >
>>> On 26.02.13 at 17:40, Michal Fiala <fiala@mfiala.net> wrote: > On 02/26/2013 05:33 PM, Jan Beulich wrote: >>>>> On 26.02.13 at 17:28, Michal Fiala <fiala@mfiala.net> wrote: >> >> Please don''t top post. >> >>> The kernel command line from grub.conf >>> >>> kernel /xen.gz dom0_mem=4G,max:4G dom0_max_vcpus=4 dom0_vcpus_pin >>> console=com2,vga com2=115200,8n1 >>> module /linux-3.7.5-hardened root=/dev/sda6 console=tty0 console=hvc0 >>> >>> I have not modified reboot type. >> >> And did you check your full boot log for an info-level message like >> >> "%s series board detected. Selecting %s-method for reboot.\n" > > I have only checked normal boot log, see attachment in my first mail. > How can I enable info-level message?Looks like this has info level messages enabled, and I can''t see any such message there. So I guess you''ll have to add a few printk()-s to arch/x86/kernel/reboot.c:native_machine_emergency_restart(). Jan
Konrad Rzeszutek Wilk
2013-Feb-26 16:56 UTC
Re: Xen 4.2.1, linux kernel 3.7.5 crashed while reboot
On Tue, Feb 26, 2013 at 03:32:43PM +0100, Michal Fiala wrote:> Hallo, > > I have updated system to xen 4.2.1, linux kernel 3.7.5. System boots > without problem. When I run reboot or shutdown -r now, system correctly > shutdowns all runlevels and then system crashed, see crash_output.txt. > I have the same problem with kernel linux 3.2.37. When I shudown system > (halt), then it is without problem. I have tested kernel 3.7.5 without > xen hypervisor, it was without problems. Required debug informations > (http://wiki.xen.org/wiki/XenParavirtOpsHelp) in attachment. > Could you please help me with this problem? > > Thanks > > Michal >> INIT: Switching to runlevel: 6 > INIT: Sending processes the TERM signal > * Stopping local > [ ok ] > * Stopping vixie-cron ... > [ ok ] > * Saving random seed ... > [ ok ] > * Deactivating additional swap space ... > [ ok ] > * Stopping sshd ... > [ ok ] > * Use of the opts variable is deprecated and will be > * removed in the future. > * Please use extra_commands, extra_started_commands or extra_stopped_commands. > * Stopping Xen control daemon ... > [ ok ] > * Stoping xenconsoled daemon ... > [ ok ] > * Stopping... ... > [ ok ] > * Stopping ntpd ... > [ ok ] > * Stopping xenstored daemon ... > [ ok ] > * Stopping nrpe ... > [ ok ] > * Unmounting network filesystems ... > [ ok ] > * Stopping munin-node ... > [ ok ] > * Stopping fail2ban ... > [ ok ] > * Stopping bacula file daemon ... > [ ok ] > * Stopping syslog-ng ... > [ ok ] > * Bringing down interface br0 > * Destroying bridge br0 ... > [ ok ] > * Bringing down interface bond0 > * Removing slaves from bond0 ... > * eth0 eth1 > [ ok ] > * Use of the opts variable is deprecated and will be > * removed in the future. > * Please use extra_commands, extra_started_commands or extra_stopped_commands. > * Stopping firewall ... > [ ok ] > * Bringing down interface lo > * Unmounting loop devices > * Unmounting filesystems > * Unmounting /boot ... > [ ok ] > * Unmounting /data/d0700 ... > [ ok ] > * Unmounting /var ... > [ ok ] > * Deactivating swap devices ... > [ ok ] > * Shutting down the Logical Volume Manager > * Shutting Down LVs & VGs ... > Command failed with status code 2. > No such command. Try ''help''. > * Failed > [ !! ] > * Finished Shutting down the Logical Volume Manager > * Stopping udev ... > [ ok ] > * Setting hardware clock using the system clock [UTC] ... > [ ok ] > * Terminating remaining processes ... > [ ok ] > * Killing remaining processes ... > [ ok ] > * Saving dependency cache ... > [ ok ] > * Remounting remaining filesystems read-only ... > * Remounting / read only ... > [ ok ] > [ ok ] > [ 612.749666] Restarting system. > [ 618.364165] int3: 0000 [#1] SMP > [ 618.364396] Modules linked in: bonding > [ 618.364590] CPU 0 > [ 618.364658] Pid: 9998, comm: reboot Tainted: G W 3.7.5-hardened #4 Dell Inc. PowerEdge R720xd/0VWT90 > [ 618.364806] RIP: e030:[<ffffffff8102a894>] [<ffffffff8102a894>] native_machine_emergency_restart+0x74/0x250 > [ 618.364956] RSP: e02b:ffff880136b35da8 EFLAGS: 00000046 > [ 618.365130] RAX: 0000000000000000 RBX: 00000000fffffffe RCX: ffffffff8172ad90 > [ 618.365308] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffffffff81ad8cdc > [ 618.365487] RBP: 0000000000000001 R08: 0000000000000000 R09: 0000000000000000 > [ 618.365663] R10: 00000000000005dd R11: 0000000000000000 R12: 0000000000000cf9 > [ 618.365867] R13: ffffffff8172ad90 R14: 0000000000000061 R15: 0000000000000000 > [ 618.366045] FS: 00007f32051f0700(0000) GS:ffff88013d600000(0000) knlGS:0000000000000000 > [ 618.366317] CS: e033 DS: 0000 ES: 0000 CR0: 0000000080050033 > [ 618.366489] CR2: 00007fa697fb3d30 CR3: 0000000135590000 CR4: 0000000000042660 > [ 618.366664] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 > [ 618.366839] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 > [ 618.367013] Process reboot (pid: 9998, threadinfo ffff880137af3b90, task ffff880137af3780) > [ 618.367288] Stack: > [ 618.367482] 0000000000000009 0000000081045315 0000000000000000 0000000001234567 > [ 618.368001] 0000000028121969 0000000000000022 0000000000000000 0000000000000000 > [ 618.368527] 0000000001234567 0000000028121969 0000000000000022 00007f3205203248 > [ 618.369079] Call Trace: > [ 618.369250] [<ffffffff8102aa9d>] ? native_machine_restart+0x2d/0x40native machine restart? It should have been xen_restart! Can you attach the full bootup serial log please? What distro is this?
On 02/26/2013 05:56 PM, Konrad Rzeszutek Wilk wrote:> On Tue, Feb 26, 2013 at 03:32:43PM +0100, Michal Fiala wrote: >> Hallo, >> >> I have updated system to xen 4.2.1, linux kernel 3.7.5. System boots >> without problem. When I run reboot or shutdown -r now, system correctly >> shutdowns all runlevels and then system crashed, see crash_output.txt. >> I have the same problem with kernel linux 3.2.37. When I shudown system >> (halt), then it is without problem. I have tested kernel 3.7.5 without >> xen hypervisor, it was without problems. Required debug informations >> (http://wiki.xen.org/wiki/XenParavirtOpsHelp) in attachment. >> Could you please help me with this problem? >> >> Thanks >> >> Michal >> > > >> INIT: Switching to runlevel: 6 >> INIT: Sending processes the TERM signal >> * Stopping local >> [ ok ] >> * Stopping vixie-cron ... >> [ ok ] >> * Saving random seed ... >> [ ok ] >> * Deactivating additional swap space ... >> [ ok ] >> * Stopping sshd ... >> [ ok ] >> * Use of the opts variable is deprecated and will be >> * removed in the future. >> * Please use extra_commands, extra_started_commands or extra_stopped_commands. >> * Stopping Xen control daemon ... >> [ ok ] >> * Stoping xenconsoled daemon ... >> [ ok ] >> * Stopping... ... >> [ ok ] >> * Stopping ntpd ... >> [ ok ] >> * Stopping xenstored daemon ... >> [ ok ] >> * Stopping nrpe ... >> [ ok ] >> * Unmounting network filesystems ... >> [ ok ] >> * Stopping munin-node ... >> [ ok ] >> * Stopping fail2ban ... >> [ ok ] >> * Stopping bacula file daemon ... >> [ ok ] >> * Stopping syslog-ng ... >> [ ok ] >> * Bringing down interface br0 >> * Destroying bridge br0 ... >> [ ok ] >> * Bringing down interface bond0 >> * Removing slaves from bond0 ... >> * eth0 eth1 >> [ ok ] >> * Use of the opts variable is deprecated and will be >> * removed in the future. >> * Please use extra_commands, extra_started_commands or extra_stopped_commands. >> * Stopping firewall ... >> [ ok ] >> * Bringing down interface lo >> * Unmounting loop devices >> * Unmounting filesystems >> * Unmounting /boot ... >> [ ok ] >> * Unmounting /data/d0700 ... >> [ ok ] >> * Unmounting /var ... >> [ ok ] >> * Deactivating swap devices ... >> [ ok ] >> * Shutting down the Logical Volume Manager >> * Shutting Down LVs & VGs ... >> Command failed with status code 2. >> No such command. Try ''help''. >> * Failed >> [ !! ] >> * Finished Shutting down the Logical Volume Manager >> * Stopping udev ... >> [ ok ] >> * Setting hardware clock using the system clock [UTC] ... >> [ ok ] >> * Terminating remaining processes ... >> [ ok ] >> * Killing remaining processes ... >> [ ok ] >> * Saving dependency cache ... >> [ ok ] >> * Remounting remaining filesystems read-only ... >> * Remounting / read only ... >> [ ok ] >> [ ok ] >> [ 612.749666] Restarting system. >> [ 618.364165] int3: 0000 [#1] SMP >> [ 618.364396] Modules linked in: bonding >> [ 618.364590] CPU 0 >> [ 618.364658] Pid: 9998, comm: reboot Tainted: G W 3.7.5-hardened #4 Dell Inc. PowerEdge R720xd/0VWT90 >> [ 618.364806] RIP: e030:[<ffffffff8102a894>] [<ffffffff8102a894>] native_machine_emergency_restart+0x74/0x250 >> [ 618.364956] RSP: e02b:ffff880136b35da8 EFLAGS: 00000046 >> [ 618.365130] RAX: 0000000000000000 RBX: 00000000fffffffe RCX: ffffffff8172ad90 >> [ 618.365308] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffffffff81ad8cdc >> [ 618.365487] RBP: 0000000000000001 R08: 0000000000000000 R09: 0000000000000000 >> [ 618.365663] R10: 00000000000005dd R11: 0000000000000000 R12: 0000000000000cf9 >> [ 618.365867] R13: ffffffff8172ad90 R14: 0000000000000061 R15: 0000000000000000 >> [ 618.366045] FS: 00007f32051f0700(0000) GS:ffff88013d600000(0000) knlGS:0000000000000000 >> [ 618.366317] CS: e033 DS: 0000 ES: 0000 CR0: 0000000080050033 >> [ 618.366489] CR2: 00007fa697fb3d30 CR3: 0000000135590000 CR4: 0000000000042660 >> [ 618.366664] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 >> [ 618.366839] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 >> [ 618.367013] Process reboot (pid: 9998, threadinfo ffff880137af3b90, task ffff880137af3780) >> [ 618.367288] Stack: >> [ 618.367482] 0000000000000009 0000000081045315 0000000000000000 0000000001234567 >> [ 618.368001] 0000000028121969 0000000000000022 0000000000000000 0000000000000000 >> [ 618.368527] 0000000001234567 0000000028121969 0000000000000022 00007f3205203248 >> [ 618.369079] Call Trace: >> [ 618.369250] [<ffffffff8102aa9d>] ? native_machine_restart+0x2d/0x40 > > native machine restart? It should have been xen_restart!Boot xen.gz, dom0 kernel, login to console as a root, run reboot.> > Can you attach the full bootup serial log please?see attachment, it is also included in my first mail in tgz package.> > What distro is this?Gentoo Base System release 2.1>Now I am playing with kernel reboot parameter, I have tried reboot=b, restart also with some problems. [ 110.416591] Restarting system. (XEN) mm.c:2518:d0 Bad type (saw 7400000000000001 != exp 4000000000000000) for ) (XEN) mm.c:2846:d0 Error while installing new baseptr 102c09b (XEN) domain_crash_sync called from entry.S (XEN) Domain 0 (vcpu#0) crashed on cpu#0: (XEN) ----[ Xen-4.2.1 x86_64 debug=n Not tainted ]---- (XEN) CPU: 0 (XEN) RIP: 000b:[<0000000000099220>] (XEN) RFLAGS: 0000000000010292 EM: 1 CONTEXT: pv guest (XEN) rax: ffff880000098000 rbx: 0000000000000000 rcx: ffffffff81b08640 (XEN) rdx: 0000000000000000 rsi: 0000000000000006 rdi: 0000000000000000 (XEN) rbp: 0000000000000000 rsp: 0000000034147d98 r8: 000000000000000a (XEN) r9: 0000000000000000 r10: 00000000000005e0 r11: 00000000000005df (XEN) r12: 0000000000000cf9 r13: ffffffff8172ad90 r14: 0000000000000062 (XEN) r15: 0000000000000000 cr0: 0000000080050033 cr4: 00000000000426f0 (XEN) cr3: 00000008fd86a000 cr2: 0000000000099220 (XEN) ds: 0000 es: 0000 fs: 0000 gs: 0000 ss: e02b cs: 000b (XEN) Guest stack trace from rsp=0000000034147d98: (XEN) Fault while accessing guest memory. (XEN) Domain 0 crashed: rebooting machine in 5 seconds.> _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xen.org > http://lists.xen.org/xen-devel >_______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
On 02/26/2013 05:51 PM, Jan Beulich wrote:>>>> On 26.02.13 at 17:40, Michal Fiala <fiala@mfiala.net> wrote: >> On 02/26/2013 05:33 PM, Jan Beulich wrote: >>>>>> On 26.02.13 at 17:28, Michal Fiala <fiala@mfiala.net> wrote: >>> >>> Please don''t top post. >>> >>>> The kernel command line from grub.conf >>>> >>>> kernel /xen.gz dom0_mem=4G,max:4G dom0_max_vcpus=4 dom0_vcpus_pin >>>> console=com2,vga com2=115200,8n1 >>>> module /linux-3.7.5-hardened root=/dev/sda6 console=tty0 console=hvc0 >>>> >>>> I have not modified reboot type. >>> >>> And did you check your full boot log for an info-level message like >>> >>> "%s series board detected. Selecting %s-method for reboot.\n" >> >> I have only checked normal boot log, see attachment in my first mail. >> How can I enable info-level message? > > Looks like this has info level messages enabled, and I can''t see any > such message there. So I guess you''ll have to add a few printk()-s > to arch/x86/kernel/reboot.c:native_machine_emergency_restart(). > > JanWhat informations are you interested in? I have attached reboot.c from my kernel. Could you please patch it and I will deploy it. Michal> > > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xen.org > http://lists.xen.org/xen-devel >_______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
Konrad Rzeszutek Wilk
2013-Feb-26 22:01 UTC
Re: Xen 4.2.1, linux kernel 3.7.5 crashed while reboot
On Tue, Feb 26, 2013 at 04:51:39PM +0000, Jan Beulich wrote:> >>> On 26.02.13 at 17:40, Michal Fiala <fiala@mfiala.net> wrote: > > On 02/26/2013 05:33 PM, Jan Beulich wrote: > >>>>> On 26.02.13 at 17:28, Michal Fiala <fiala@mfiala.net> wrote: > >> > >> Please don''t top post. > >> > >>> The kernel command line from grub.conf > >>> > >>> kernel /xen.gz dom0_mem=4G,max:4G dom0_max_vcpus=4 dom0_vcpus_pin > >>> console=com2,vga com2=115200,8n1 > >>> module /linux-3.7.5-hardened root=/dev/sda6 console=tty0 console=hvc0 > >>> > >>> I have not modified reboot type. > >> > >> And did you check your full boot log for an info-level message like > >> > >> "%s series board detected. Selecting %s-method for reboot.\n" > > > > I have only checked normal boot log, see attachment in my first mail. > > How can I enable info-level message? > > Looks like this has info level messages enabled, and I can''t see any > such message there. So I guess you''ll have to add a few printk()-s > to arch/x86/kernel/reboot.c:native_machine_emergency_restart().The thing is is that native_machine_emergency_restart is not suppose to be called. The baremetal machine_ops has that, but we over-write it with our own and he should get: .emergency_restart = xen_emergency_restart, ! So how did his .emergency_restart value get over-written by native_machine_emergency_restart?> > Jan > > > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xen.org > http://lists.xen.org/xen-devel >
>>> On 26.02.13 at 23:01, Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> wrote: > The thing is is that native_machine_emergency_restart is not suppose > to be called. The baremetal machine_ops has that, but we over-write > it with our own and he should get: > > .emergency_restart = xen_emergency_restart, > > ! So how did his .emergency_restart value get over-written by > native_machine_emergency_restart?Indeed. So perhaps the -hardened tag indicates some patching having gone on that we''re unaware of. Michal - without knowing what your kernel does, I don''t think we can really help. And please remember that this is xen-devel, not xen-users, so we expect you to be able to do at least a minimal share of the work needed to find out what is going on. First step would probably be to try plain 3.7.5 (or subsequent 3.7.x). Jan
On 02/27/2013 09:35 AM, Jan Beulich wrote:>>>> On 26.02.13 at 23:01, Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> wrote: >> The thing is is that native_machine_emergency_restart is not suppose >> to be called. The baremetal machine_ops has that, but we over-write >> it with our own and he should get: >> >> .emergency_restart = xen_emergency_restart, >> >> ! So how did his .emergency_restart value get over-written by >> native_machine_emergency_restart? > > Indeed. So perhaps the -hardened tag indicates some patching > having gone on that we''re unaware of. Michal - without knowing > what your kernel does, I don''t think we can really help. And > please remember that this is xen-devel, not xen-users, so we > expect you to be able to do at least a minimal share of the work > needed to find out what is going on. First step would probably > be to try plain 3.7.5 (or subsequent 3.7.x). > > Jan >We are using gentoo hardened sources (kernel). Hardened-sources is based on the official Linux kernel and is targeted at our users running Gentoo on server systems. It provides patches for the various subprojects of Gentoo Hardened (such as support for LSM/SELinux and grsecurity), together with stability/security-enhancements. See http://www.gentoo.org/doc/en/gentoo-kernel.xml I will try gentoo-sources (lightly patched to fix security problems, kernel bugs, and to increase compatibility with the more uncommon system architectures) and vanilla-sources (official kernel sources released on http://www.kernel.org/) Thanks Michal
Konrad Rzeszutek Wilk
2013-Feb-27 15:41 UTC
Re: Xen 4.2.1, linux kernel 3.7.5 crashed while reboot
On Wed, Feb 27, 2013 at 11:08:19AM +0100, Michal Fiala wrote:> On 02/27/2013 09:35 AM, Jan Beulich wrote: > >>>> On 26.02.13 at 23:01, Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> wrote: > >> The thing is is that native_machine_emergency_restart is not suppose > >> to be called. The baremetal machine_ops has that, but we over-write > >> it with our own and he should get: > >> > >> .emergency_restart = xen_emergency_restart, > >> > >> ! So how did his .emergency_restart value get over-written by > >> native_machine_emergency_restart? > > > > Indeed. So perhaps the -hardened tag indicates some patching > > having gone on that we''re unaware of. Michal - without knowing > > what your kernel does, I don''t think we can really help. And > > please remember that this is xen-devel, not xen-users, so we > > expect you to be able to do at least a minimal share of the work > > needed to find out what is going on. First step would probably > > be to try plain 3.7.5 (or subsequent 3.7.x). > > > > Jan > > > > We are using gentoo hardened sources (kernel). Hardened-sources is based > on the official Linux kernel and is targeted at our users running Gentoo > on server systems. It provides patches for the various subprojects of > Gentoo Hardened (such as support for LSM/SELinux and grsecurity), > together with stability/security-enhancements. > See http://www.gentoo.org/doc/en/gentoo-kernel.xml > > I will try gentoo-sources (lightly patched to fix security problems, > kernel bugs, and to increase compatibility with the more uncommon system > architectures) and vanilla-sources (official kernel sources released on > http://www.kernel.org/)Excellent. Thanks.> > Thanks > > Michal > > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xen.org > http://lists.xen.org/xen-devel >
On 02/27/2013 04:41 PM, Konrad Rzeszutek Wilk wrote:> On Wed, Feb 27, 2013 at 11:08:19AM +0100, Michal Fiala wrote: >> On 02/27/2013 09:35 AM, Jan Beulich wrote: >>>>>> On 26.02.13 at 23:01, Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> wrote: >>>> The thing is is that native_machine_emergency_restart is not suppose >>>> to be called. The baremetal machine_ops has that, but we over-write >>>> it with our own and he should get: >>>> >>>> .emergency_restart = xen_emergency_restart, >>>> >>>> ! So how did his .emergency_restart value get over-written by >>>> native_machine_emergency_restart? >>> >>> Indeed. So perhaps the -hardened tag indicates some patching >>> having gone on that we''re unaware of. Michal - without knowing >>> what your kernel does, I don''t think we can really help. And >>> please remember that this is xen-devel, not xen-users, so we >>> expect you to be able to do at least a minimal share of the work >>> needed to find out what is going on. First step would probably >>> be to try plain 3.7.5 (or subsequent 3.7.x). >>> >>> Jan >>> >> >> We are using gentoo hardened sources (kernel). Hardened-sources is based >> on the official Linux kernel and is targeted at our users running Gentoo >> on server systems. It provides patches for the various subprojects of >> Gentoo Hardened (such as support for LSM/SELinux and grsecurity), >> together with stability/security-enhancements. >> See http://www.gentoo.org/doc/en/gentoo-kernel.xml >> >> I will try gentoo-sources (lightly patched to fix security problems, >> kernel bugs, and to increase compatibility with the more uncommon system >> architectures) and vanilla-sources (official kernel sources released on >> http://www.kernel.org/) > > Excellent. Thanks.The problem was fixed by changing kernel to gentoo-sources. So there is a problem in hardened-sources. Thanks M.>> >> Thanks >> >> Michal >> >> _______________________________________________ >> Xen-devel mailing list >> Xen-devel@lists.xen.org >> http://lists.xen.org/xen-devel >> > > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xen.org > http://lists.xen.org/xen-devel >
Roger Pau Monné
2013-Mar-04 16:44 UTC
Re: Xen 4.2.1, linux kernel 3.7.5 crashed while reboot
On 04/03/13 17:37, Michal Fiala wrote:>>> We are using gentoo hardened sources (kernel). Hardened-sources is based >>> on the official Linux kernel and is targeted at our users running Gentoo >>> on server systems. It provides patches for the various subprojects of >>> Gentoo Hardened (such as support for LSM/SELinux and grsecurity), >>> together with stability/security-enhancements. >>> See http://www.gentoo.org/doc/en/gentoo-kernel.xml >>> >>> I will try gentoo-sources (lightly patched to fix security problems, >>> kernel bugs, and to increase compatibility with the more uncommon system >>> architectures) and vanilla-sources (official kernel sources released on >>> http://www.kernel.org/) >> >> Excellent. Thanks. > > The problem was fixed by changing kernel to gentoo-sources. So there is > a problem in hardened-sources.Alpine Linux is also using at least the grsecurity patches and there seems to be no problem with Xen.