I have just tried to increase the memory of a DomU from 512 MB to
1024MB. I have several other DomUs already running with more than
512Mb, so I am puzzled.
There is more than sufficient free RAM on the host Dom0 currently has
2.9G allocated to it. But I get the following error. If I change it
back to 512 it starts just fine.
I am running xen-3.3.0-3 on CentyOS5.2 x86_64 using the rpms from gitco.
NB: I have always had the PCI: Fatal message on all my pv DomUs.
Started domain agree-25
PCI: Fatal: No config space access function found
isapnp: Write Data Register 0xa79 already used
1 multicall(s) failed: cpu 0
------------[ cut here ]------------
kernel BUG at arch/x86/xen/multicalls.c:104!
invalid opcode: 0000 [#1] SMP
Modules linked in:
Pid: 1, comm: init Not tainted (2.6.27.9-159.fc10.i686.PAE #1)
EIP: 0061:[<c0404a97>] EFLAGS: 00010202 CPU: 0
EIP is at xen_mc_flush+0x17b/0x187
EAX: c16c5054 EBX: 00000000 ECX: ffffffff EDX: c16c5054
ESI: 00000001 EDI: 00000001 EBP: ed013ef8 ESP: ed013ee4
DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0069
Process init (pid: 1, ti=ed013000 task=ed030000 task.ti=ed013000)
Stack: c16c5054 c16c0254 00a72000 f57aa388 ec3f0000 ed013f00 c0404720
ed013f74
c04871d6 009c9065 80000000 00a71fff 00a71fff 80000000 00000025
00a70000
ed013fa0 ec3f0000 08100073 00000002 00000000 ec3d0028 ec3cf000
ec3f0000
--
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-Glen and
Rosanne Eustace,
GodZone Internet Services, a division of AGRE Enterprises Ltd.,
P.O. Box 8020, Palmerston North, New Zealand 4446
Ph/Fax: +64 6 357 8168, Mob: +64 27 542 4015
"A Ministry specialising in providing low-cost professional Internet
Services to NZ Christian Churches, Ministries and Organisations"
_______________________________________________
Xen-users mailing list
Xen-users@lists.xensource.com
http://lists.xensource.com/xen-users
I probably needed to add all of the following to show why I think this is weird. I have 13 DomUs on this server, 1x Win2k3 HVM, 1x Fedora 8 and all the rest are Fedora 10, all running the same kernel 2.6.27.9-159.fc10.i686.PAE, the F10 DomUs have a mixture of 1 to 4 VCPUs and memory footprints from 128 - 1024 Mb. I built another DomU from the same template last night and set it to run with 1024 Mb and it ran just fine, no kernel panic. So, my question is, why does a DomU that is pretty much the same as all of the others panic when I change its memory parameters. NB: I am shutting down the DomU, updating the /etc/xen conf file and then doing an xm create. Boom And if I return to what it was it goes again. It their some residual data in the hypervisor or something ? -- =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-Glen and Rosanne Eustace, GodZone Internet Services, a division of AGRE Enterprises Ltd., P.O. Box 8020, Palmerston North, New Zealand 4446 Ph/Fax: +64 6 357 8168, Mob: +64 27 542 4015 "A Ministry specialising in providing low-cost professional Internet Services to NZ Christian Churches, Ministries and Organisations" _______________________________________________ Xen-users mailing list Xen-users@lists.xensource.com http://lists.xensource.com/xen-users