Hi Simon,
I can confirm that the cpu mode is host-passthrough
FYI - I run KVM for my Hypervisor
If this is any help I took a screenshot of the console.
6.1522631 Call Trace:
6.1525591
6.1528651
dump_stack_lvl+0x34/0x48
panic+0x102/0x2d4
6.153139]
6.153438]
do_exit.cold+0x14/0x9f
do_group_exit+0x33/Oxa0
6.1537371
6.1540621
6.154393]
? exc_page_fault+0x62/0x150
6.1547091
entry_SYSCALL_64_after_hwframe+0x63/Oxcd
6.1550791 RIP: 0033:0x78 acobb?c2d
6.1554101 Code: c3 Of 11 84 00 00 00 00 00 f3 Of 1e fa be e7 00 00 00 ba
3cl
00
00 00 eb od 89 do or 05 48 3d 00 fo ff ff 27 1c f4 89 fo of 05 <48› 3d 00 fo
ff
ff
76 e7 f7 d8 89 05 ff fe 00 00 eb dd of 1f
44 00
6.1565621 RSP: 002h: 00007ffcf88f6428 EFLAGS: 00000246 ORIG RAX: 00000000000
000e7
6.1570511 RAX: fffffffffffffida RBX: 00007 dc0bb76130 RCX: 00007fdc0bb7c2d1
6.1575451 RDX: 000000000000003c RSI: 00000000000000€7 RDI: 000000000000007f
6.1580151 RB: 00007ffcf88f6eb0 R08: 00007ffcf8816899 R09: 0000000000000000
6.1586231 R10: 00000000ffffffff R11: 0000000000000246 R12: 00007f dcObb55000
6.1591971 R13: 0000002300000007 R14: 0000000000000007 R15: 00007ffcf88f6ec0
6.1617161 Kernel Offset: 0x2fc00000 from Oxffffffff81000000 (relocation ran
ge: Oxffffffff80000000-Oxffffffffbfffffff)
6.1626471
-[ end Kernel panic
- not syncing: Attempted to kill init? exit
code=0x00007f 00
Thanks
On 2/2/23, 3:11 PM, "Simon Weller" <siwelle...@gmail.com> wrote:
Jeremy and Robert,
Out of interest, which cpu profile are you passing through to the guest?
I've seen a lot of reports with 9.x and kernel panics when certain CPU
flags aren't being passed through, in particular x16 lahf popcnt sse4_1
sse4_2 ssse3.
Do you have the ability to set your guest.cpu.mode to "host-passthrough"
for the purposes of testing?
You'd do that by editing /etc/cloudstack/agent/agent.properties on your
host.
Look for guest.cpu.mode and replace it with =
guest.cpu.mode=host-passthrough
Then restart the agent.
-Si
On Thu, Feb 2, 2023 at 9:32 AM Robert Ward <rww...@xcubia.com> wrote:
> Hi Rohit,
>
> I can report that installing Oracle 9.1 on a VM results in a kernel panic
> right off the bat. I hear there were some major changes between EL8 and 9.
> My SaSS customers are holding off on upgrades as well because of potential
> concerns.
>
> Thanks
>
> On 2/2/23, 10:17 AM, "Rohit Yadav" <rohit.ya...@shapeblue.com> wrote:
>
> Jeremy,
>
> For EL9, can you try Alma Linux 9 or Oracle Linux 9 for guest VM? I
> have had bad experiences with Rocky Linux 9 in terms of stability and
> mirror/repo accessibility even though Rocky Linux 9 claims to be EL9
> compatible.
>
> Regards.
>
> ________________________________
> From: Jeremy Hansen <jer...@skidrow.la.INVALID>
> Sent: Saturday, December 10, 2022 7:53 PM
> To: Vivek Kumar via users <users@cloudstack.apache.org>
> Subject: Rocky 9 and CS 4.17.1.0
>
> I’m running Cloudstack 4.17.1.0 and for unknown reasons, I’m having
> issues running Rocky 9. Kernel begins to boot and then it looks like it
> fails on loading initrd and I get a kernel oops. Just curious if this is
a
> known issue or if there’s a work around. I tried using the qcow2 image
from
> Rocky as well and just using the install iso to create a new image. Same
> result.
>
> Rocky 8 works fine.
>
> Anyone running Rocky 9?
>
> Thanks
> -jeremy
>
>
>
>
>
>
>
>
>
