I''m trying to install a paravirtualized instance of build 127 (domU) on a machine running CentOS 5.4 (dom0). I''m able to start the domU and complete approximately 98% of the installation. During the final part of the installation, the dom0 apparently panics. I say "apparently" because I haven''t been able to find anything that indicates a panic occurred; the machine simply reboots. There aren''t any errors or warnings logged in "/var/log/messages" or in "/var/log/xen/xend.log", and I do not see any messages on the console about a panic. I also set up kdump on the dom0 and it does not produce a crash dump when this problem occurs. I''ve searched the Bugzilla repo for both Opensolaris and Red Hat Linux, but I haven''t found any bugs with similar symptoms. Anyone have any ideas on why this might happen or what I could do to try and determine the cause? -- This message posted from opensolaris.org
On Mon, Nov 30, 2009 at 04:46:59PM -0800, Andrew Chace wrote:> I''m trying to install a paravirtualized instance of build 127 (domU) > on a machine running CentOS 5.4 (dom0). I''m able to start the domU and > complete approximately 98% of the installation. During the final part > of the installation, the dom0 apparently panics. I say "apparently" > because I haven''t been able to find anything that indicates a panic > occurred; the machine simply reboots. There aren''t any errors or > warnings logged in "/var/log/messages" or in "/var/log/xen/xend.log", > and I do not see any messages on the console about a panic. I also set > up kdump on the dom0 and it does not produce a crash dump when this > problem occurs.Asking at xen-users@lists.xensource.com is your best bet - we don''t work on Linux dom0 :) regards john
Point taken ;-). I''ll post the question to xen-users, and report back if I resolve the problem. Can I assume that there isn''t anything obviously/glaringly wrong with trying to run a snv_127 on top of CentOS 5.4? Is there a minimum version of the hypervisor required to run snv_127? -- This message posted from opensolaris.org
> During the final part of the installation, the dom0 apparently panicsI would suggest to setup serial console, if panics mean Dom0 crash. Just forwarding you statement to xen-users or xen-devel won''t help much. Any xenified kernel trace ( if it happens at all) would go to serial output. Could you also post "xm info" ? -- This message posted from opensolaris.org
>the machine simply rebootsI missed it. Serial console report is a must to get help. Kernel trace would be there. -- This message posted from opensolaris.org
View :- http://opensolaris.org/jive/thread.jspa?messageID=439445 Most probably Dom0 kernel issue. -- This message posted from opensolaris.org
Here''s the output from "xm info": [root@centos xen]# xm info host : centos.network release : 2.6.18-164.el5xen version : #1 SMP Thu Sep 3 04:47:32 EDT 2009 machine : i686 nr_cpus : 1 nr_nodes : 1 sockets_per_node : 1 cores_per_socket : 1 threads_per_core : 1 cpu_mhz : 1794 hw_caps : 3febfbff:00000000:00000000:00000080 total_memory : 1015 free_memory : 0 node_to_cpu : node0:0 xen_major : 3 xen_minor : 1 xen_extra : .2-164.el5 xen_caps : xen-3.0-x86_32p xen_pagesize : 4096 platform_params : virt_start=0xf5800000 xen_changeset : unavailable cc_compiler : gcc version 4.1.2 20080704 (Red Hat 4.1.2-46) cc_compile_by : mockbuild cc_compile_domain : centos.org cc_compile_date : Thu Sep 3 03:26:12 EDT 2009 xend_config_format : 2 I''ll post back after I''ve set up the serial console. -- This message posted from opensolaris.org
I was able to install OSOL 1002-127 in PV DomU at Xen 3.4.2 Dom0 on top of CentOS 5.3. Hypervisor upgraded via http://gitco.de/repo. Just to keep you posted , ZFS 16 support and patch for pygrub came in official Xen 3.4.1 release ( maybe 3.4.0). My environment :- [root@RHEL53 ~]# xm info host : RHEL53 release : 2.6.18-164.6.1.el5xen version : #1 SMP Tue Nov 3 16:48:13 EST 2009 machine : x86_64 nr_cpus : 2 nr_nodes : 1 cores_per_socket : 2 threads_per_core : 1 cpu_mhz : 2999 hw_caps : bfebfbff:20100800:00000000:00000140:0008e3fd:00000000:00000001:00000000 virt_caps : hvm total_memory : 8190 free_memory : 11 node_to_cpu : node0:0-1 node_to_memory : node0:11 xen_major : 3 xen_minor : 4 xen_extra : .2 xen_caps : xen-3.0-x86_64 xen-3.0-x86_32p hvm-3.0-x86_32 hvm-3.0-x86_32p hvm-3.0-x86_64 xen_scheduler : credit xen_pagesize : 4096 platform_params : virt_start=0xffff800000000000 xen_changeset : unavailable cc_compiler : gcc version 4.1.2 20080704 (Red Hat 4.1.2-44) cc_compile_by : root cc_compile_domain : cc_compile_date : Sun Nov 29 15:23:49 MSK 2009 xend_config_format : 4 -- This message posted from opensolaris.org
I believe you do :- http://blogs.sun.com/levon/entry/opensolaris_2008_11_guest_domain -- This message posted from opensolaris.org
I''ve added the 32-bit gitco.de repository to my yum config; in the process of updating Xen packages right now. The latest hypervisor available in the 32-bit repo is version 3.4.0; hopefully it contains support for ZFS and pygrub. I''ll post back after I''ve tried to install OSOL 1002-127. -- This message posted from opensolaris.org
I updated to Xen 3.4.0 via the gitco.de repository. I''ve been able to reproduce the "crash" twice (out of two attempts to install OSOL 1002-127). Here''s output from "xm info" after the upgrade: [root@centos ajchace]# xm info host : centos.network release : 2.6.18-164.6.1.el5xen version : #1 SMP Tue Nov 3 17:53:47 EST 2009 machine : i686 nr_cpus : 1 nr_nodes : 1 cores_per_socket : 1 threads_per_core : 1 cpu_mhz : 1794 hw_caps : 3febfbff:00000000:00000000:00000080:00000000:00000000:00000000:00000000 virt_caps : total_memory : 1015 free_memory : 62 node_to_cpu : node0:0 node_to_memory : node0:62 xen_major : 3 xen_minor : 4 xen_extra : .0 xen_caps : xen-3.0-x86_32p xen_scheduler : credit xen_pagesize : 4096 platform_params : virt_start=0xf5800000 xen_changeset : unavailable cc_compiler : gcc version 4.1.2 20071124 (Red Hat 4.1.2-42) cc_compile_by : root cc_compile_domain : gitco.tld cc_compile_date : Sat May 30 17:47:40 CEST 2009 I''ll try to get a serial cable made for the console tomorrow. -- This message posted from opensolaris.org
If 32-bit architecture is due to hardware limitations. It might be very hard to obtain support, even via RH''s Bugzilla report with serial log attached. If it''s just 32-bit OS ( CentOS 5.4) installed on 64-bit box ( Core2Duo, Core2Quad and etc) reinstall 64-bit OS would be the shortest way. -- This message posted from opensolaris.org