SYSTEM WARNING: 'date_default_timezone_get(): It is not safe to rely on the system's timezone settings. You are *required* to use the date.timezone setting or the date_default_timezone_set() function. In case you used any of those methods and you are still getting this warning, you most likely misspelled the timezone identifier. We selected the timezone 'UTC' for now, but please set date.timezone to select your timezone.' in '/usr/share/mantis/www/core.php' line 264

0002982: 2.4 has no PAE kernel for domU - MantisBT
MantisBT - Endian Firewall
View Issue Details
0002982Endian FirewallXenpublic2010-06-07 07:192010-06-09 14:42
aender 
ra-endian 
normalmajoralways
closedfixed 
2.4 
2.42.4 
0002982: 2.4 has no PAE kernel for domU
The Endian 2.4 kernel is not compiled with pae. So it can´t run as an domU on Xen.

Look at this: http://wiki.xensource.com/xenwiki/XenParavirtOps [^]

Building with domU support

   1. Get a current kernel. The latest kernel.org kernel is generally a good choice.
   2. Configure as normal; you can start with your current .config file
   3. If building 32 bit kernel make sure you have CONFIG_X86_PAE enabled (which is set by selecting CONFIG_HIGHMEM64G)
          * non-PAE mode doesn't work in 2.6.25, and has been dropped altogether from 2.6.26 and newer kernel versions.
   4. Enable these core options:
         1. CONFIG_PARAVIRT_GUEST
         2. CONFIG_XEN
   5. And Xen pv device support
         1. CONFIG_HVC_DRIVER and CONFIG_HVC_XEN
         2. CONFIG_XEN_BLKDEV_FRONTEND
         3. CONFIG_XEN_NETDEV_FRONTEND
   6. And build as usual
No tags attached.
Issue History
2010-06-07 07:19aenderNew Issue
2010-06-07 08:22peter-endianStatusnew => confirmed
2010-06-07 14:14peter-endianTarget Version => 2.4
2010-06-07 14:23peter-endianStatusconfirmed => new
2010-06-07 14:23peter-endianAssigned To => ra-endian
2010-06-07 14:23peter-endianStatusnew => assigned
2010-06-08 09:55ra-endianNote Added: 0004426
2010-06-08 09:55ra-endianStatusassigned => closed
2010-06-08 09:55ra-endianResolutionopen => fixed
2010-06-08 09:55ra-endianFixed in Version => 2.4
2010-06-08 11:32aenderNote Added: 0004432
2010-06-08 11:32aenderStatusclosed => feedback
2010-06-08 11:32aenderResolutionfixed => reopened
2010-06-08 12:47ra-endianNote Added: 0004435
2010-06-08 12:47ra-endianStatusfeedback => closed
2010-06-08 12:47ra-endianResolutionreopened => fixed
2010-06-08 13:02christian-endianNote Edited: 0004435
2010-06-08 13:30aenderNote Added: 0004437
2010-06-08 13:30aenderStatusclosed => feedback
2010-06-08 13:30aenderResolutionfixed => reopened
2010-06-08 13:39christian-endianNote Added: 0004438
2010-06-08 13:39christian-endianStatusfeedback => closed
2010-06-08 13:39christian-endianResolutionreopened => fixed
2010-06-09 11:44aenderNote Added: 0004454
2010-06-09 11:44aenderStatusclosed => feedback
2010-06-09 11:44aenderResolutionfixed => reopened
2010-06-09 13:42christian-endianNote Added: 0004457
2010-06-09 13:44aenderNote Added: 0004458
2010-06-09 13:45aenderStatusfeedback => resolved
2010-06-09 13:45aenderResolutionreopened => fixed
2010-06-09 14:42christian-endianStatusresolved => closed

Notes
(0004426)
ra-endian   
2010-06-08 09:55   
released into the community repo.

you can install the package with:
smart install kernel-PAE
(0004432)
aender   
2010-06-08 11:32   
You forgot to build the xen modules.

Look at point 5.

5. And Xen pv device support
         1. CONFIG_HVC_DRIVER and CONFIG_HVC_XEN
         2. CONFIG_XEN_BLKDEV_FRONTEND
         3. CONFIG_XEN_NETDEV_FRONTEND
(0004435)
ra-endian   
2010-06-08 12:47   
(edited on: 2010-06-08 13:02)
the current PAE kernel has all XEN Features enabled:

CONFIG_XEN=y
CONFIG_XEN_MAX_DOMAIN_MEMORY=8
CONFIG_XEN_SAVE_RESTORE=y
CONFIG_XEN_BLKDEV_FRONTEND=y
CONFIG_XEN_NETDEV_FRONTEND=y
CONFIG_XEN_KBDDEV_FRONTEND=y
CONFIG_XEN_FBDEV_FRONTEND=y
CONFIG_XEN_BALLOON=y
CONFIG_XEN_SCRUB_PAGES=y
CONFIG_PARAVIRT_GUEST=y
CONFIG_PARAVIRT=y
CONFIG_PARAVIRT_CLOCK=y
CONFIG_HVC_DRIVER=y
CONFIG_HVC_XEN=y

(0004437)
aender   
2010-06-08 13:30   
Sorry.

My fault. Machine crashed because we have 786 MB memory.
With 512, 896 or 1024 MB it works fine.

We have the same issue with a fedora kernel on a other xen installation.

Here the log:


[ 0.058074] PCI: Fatal: No config space access function found
[ 0.078411] isapnp: Write Data Register 0xa79 already used
[ 0.118877] i8042.c: No controller found.
[ 0.131222] ide_generic: I/O resource 0x1F0-0x1F7 not free.
[ 0.131230] ide_generic: I/O resource 0x170-0x177 not free.
[ 0.232785] 1 multicall(s) failed: cpu 0
[ 0.233108] ------------[ cut here ]------------
[ 0.233121] kernel BUG at arch/x86/xen/multicalls.c:104!
[ 0.233135] invalid opcode: 0000 [0000001] SMP
[ 0.233154] Modules linked in:
[ 0.233168]
[ 0.233177] Pid: 1, comm: init Not tainted (2.6.27.19-72.e25PAE 0000001)
[ 0.233191] EIP: 0061:[<c1004716>] EFLAGS: 00010202 CPU: 0
[ 0.233205] EIP is at xen_mc_flush+0x136/0x140
[ 0.233217] EAX: c1380260 EBX: c1b7da60 ECX: 00000000 EDX: c13f008c
[ 0.233230] ESI: 00000000 EDI: 00000001 EBP: c1b7d060 ESP: ed013d04
[ 0.233244] DS: 007b ES: 007b FS: 00d8 GS: 0000 SS: e021
[ 0.233257] Process init (pid: 1, ti=ed013000 task=ed030000 task.ti=ed013000)
[ 0.233271] Stack: c1380260 00000001 00000001 00000000 c154f000 c106ecbb ed013d2c ed013d2c
[ 0.233319] ed04e280 00000000 00000001 00000000 c0000000 ec956000 ec94a000 00100173
[ 0.233365] 00000001 00000000 ec81d010 ec94a000 f57b4000 c1b67a8c ec954ff8 00000025
[ 0.233412] Call Trace:
[ 0.233423] [<c106ecbb>] mprotect_fixup+0x47f/0x5ee
[ 0.233441] [<c1083ec3>] setup_arg_pages+0xc0/0x244
[ 0.233458] [<c10a791e>] load_elf_binary+0x3df/0x10b3
[ 0.233475] [<c107ed90>] __dentry_open+0x130/0x1fc
[ 0.233493] [<c1080a72>] vfs_read+0xe3/0x11e
[ 0.233510] [<c10838ad>] search_binary_handler+0x9b/0x202
[ 0.233528] [<c10a6ecd>] load_script+0x191/0x1a4
[ 0.233545] [<c1005e3e>] check_events+0x8/0xe
[ 0.233561] [<c1139004>] force_evtchn_callback+0xc/0x10
[ 0.233579] [<c1005e3e>] check_events+0x8/0xe
[ 0.233596] [<c1005da3>] xen_restore_fl_direct_end+0x0/0x1
[ 0.233614] [<c12358d5>] _spin_unlock_irqrestore+0x13/0x17
[ 0.233633] [<c106659c>] page_address+0x73/0x93
[ 0.233650] [<c10837dc>] copy_strings+0x169/0x173
[ 0.233667] [<c10838ad>] search_binary_handler+0x9b/0x202
[ 0.233685] [<c1084aad>] do_execve+0x149/0x1f2
[ 0.233702] [<c100712d>] sys_execve+0x2a/0x49
[ 0.233719] [<c1008a5e>] system_call_done+0x0/0x4
[ 0.233736] [<c107007b>] try_to_unmap_one+0x5f/0x235
[ 0.233754] [<c100b5cc>] kernel_execve+0x14/0x18
[ 0.233770] [<c10031b8>] init_post+0x98/0xfc
[ 0.233787] [<c100969f>] kernel_thread_helper+0x7/0x10
[ 0.233803] =======================
[ 0.233813] Code: 5a 59 31 f6 8d 9d 00 0a 00 00 eb 09 8b 43 04 46 ff 13 83 c3 08 3b b5 08 0b 00 00 72 ef 85 ff c7 85 08 0b 00 00 00 00 00 00 74 04 <0f> 0b eb fe 5b 5b 5e 5f 5d c3 57 89 d7 56 89 c6 53 b8 60 50 38
[ 0.234091] EIP: [<c1004716>] xen_mc_flush+0x136/0x140 SS:ESP e021:ed013d04
[ 0.234091] ---[ end trace 4b88ae1b1dc6700a ]---
[ 0.234091] note: init[1] exited with preempt_count 1
[ 0.234162] BUG: scheduling while atomic: init/1/0x00000001
[ 0.234174] Modules linked in:
[ 0.234189] Pid: 1, comm: init Tainted: G D 2.6.27.19-72.e25PAE 0000001
[ 0.234204] [<c1233e24>] schedule+0x71/0x66a
[ 0.234220] [<c100ab77>] do_IRQ+0x99/0xae
[ 0.234237] [<c12356e4>] rwsem_down_failed_common+0x153/0x16e
[ 0.234256] [<c1235738>] rwsem_down_read_failed+0x1a/0x22
[ 0.235154] [<c1235773>] call_rwsem_down_read_failed+0x7/0xc
[ 0.235154] [<c1234d30>] down_read+0x14/0x16
[ 0.235154] [<c102e512>] exit_mm+0x28/0xe5
[ 0.235154] [<c1005d6b>] xen_irq_enable_direct_end+0x0/0x1
[ 0.235154] [<c114e804>] tty_audit_exit+0x53/0x7d
[ 0.235154] [<c102f9a4>] do_exit+0x1a8/0x6a3
[ 0.235154] [<c1235d1c>] oops_end+0x9a/0x9e
[ 0.235154] [<c1235f13>] do_trap+0xaf/0xe3
[ 0.235154] [<c1009e15>] do_invalid_op+0x0/0x6b
[ 0.235154] [<c1009e77>] do_invalid_op+0x62/0x6b
[ 0.235154] [<c1004716>] xen_mc_flush+0x136/0x140
[ 0.235154] [<c1005da3>] xen_restore_fl_direct_end+0x0/0x1
[ 0.235154] [<c100aad7>] do_softirq+0xa2/0xa9
[ 0.235154] [<c1056c42>] handle_level_irq+0x0/0xc2
[ 0.235154] [<c1031266>] irq_exit+0x50/0x69
[ 0.235154] [<c100ab77>] do_IRQ+0x99/0xae
[ 0.235154] [<c1139b4c>] xen_evtchn_do_upcall+0xc1/0xf1
[ 0.235154] [<c10096ef>] xen_do_upcall+0x7/0xc
[ 0.235154] [<c1235ada>] error_code+0x72/0x78
[ 0.235154] [<c1004716>] xen_mc_flush+0x136/0x140
[ 0.235154] [<c106ecbb>] mprotect_fixup+0x47f/0x5ee
[ 0.235154] [<c1083ec3>] setup_arg_pages+0xc0/0x244
[ 0.235154] [<c10a791e>] load_elf_binary+0x3df/0x10b3
[ 0.235154] [<c107ed90>] __dentry_open+0x130/0x1fc
[ 0.235154] [<c1080a72>] vfs_read+0xe3/0x11e
[ 0.235154] [<c10838ad>] search_binary_handler+0x9b/0x202
[ 0.235154] [<c10a6ecd>] load_script+0x191/0x1a4
[ 0.235154] [<c1005e3e>] check_events+0x8/0xe
[ 0.235154] [<c1139004>] force_evtchn_callback+0xc/0x10
[ 0.235154] [<c1005e3e>] check_events+0x8/0xe
[ 0.235154] [<c1005da3>] xen_restore_fl_direct_end+0x0/0x1
[ 0.235154] [<c12358d5>] _spin_unlock_irqrestore+0x13/0x17
[ 0.235154] [<c106659c>] page_address+0x73/0x93
[ 0.235154] [<c10837dc>] copy_strings+0x169/0x173
[ 0.235154] [<c10838ad>] search_binary_handler+0x9b/0x202
[ 0.235154] [<c1084aad>] do_execve+0x149/0x1f2
[ 0.235154] [<c100712d>] sys_execve+0x2a/0x49
[ 0.235154] [<c1008a5e>] system_call_done+0x0/0x4
[ 0.235154] [<c107007b>] try_to_unmap_one+0x5f/0x235
[ 0.235154] [<c100b5cc>] kernel_execve+0x14/0x18
[ 0.235154] [<c10031b8>] init_post+0x98/0xfc
[ 0.235154] [<c100969f>] kernel_thread_helper+0x7/0x10
[ 0.235154] =======================
(0004438)
christian-endian   
2010-06-08 13:39   
Bug says there is no support for XEN - this has been fixed -> bug closed!
(0004454)
aender   
2010-06-09 11:44   
What do you mean with:

"Bug says there is no support for XEN - this has been fixed -> bug closed!"

The message is from your PAE Kernel starting as PV on a xen host with 786MB for the VM.
With 512, 896 or 1024 MB it works fine.

So "no support for XEN" could not be the problem!?
(0004457)
christian-endian   
2010-06-09 13:42   
The Bug summary is "2.4 has no PAE kernel for domU". Do you think this has been resolved? Good - because I think so too! Therefore I closed the bug!

Everything else is in no way related to the initial bug and should be handled in a separate bug report. There are a few reasons for this:

1. People who read only summaries still think there is no XEN domU support because the bug is still open.
2. People who happen to have the same problem as you will report this again and again because the summary of *this* bug does not tell them to look here.

Last but not least - have you tried using 768MB instead of 786MB?

On a side note I might like to add the following:
If it works with 512, 896 or 1024MB frankly I do not care if it does not work with 786MB. Most of all because this looks like a Xen-problem to me, especially since you tell me this is happening on fedora as well. I really hope you do not expect us to solve upstream problems if they do not interfere with the functionality of our software at all.

Kind regards,
Christian
(0004458)
aender   
2010-06-09 13:44   
Thanks for your response.

So we close this bug.