Re: [Qemu-devel] Slow linux (RH5.6) guest reboot?

[Kenton Cabiness <kenton.cabiness@alcatel-lucent.com>]

One thing I forgot to mention is that this same guest image has no 
problem when installed bare metal.

Kenton

On 10/3/2011 1:20 PM, Kenton Cabiness wrote:
> In general, we have seen that during a reboot of a
> running guest (by issuing a 'reboot' command in the
> guest), the guest goes down normally, but when coming
> back up, during the udev phase, it takes anywhere from
> a few seconds to a few hours to complete that part of the
> boot.  Then the guest seems to behave itself after that.
> If the guest is powered off (virsh destroy guest), then
> restarted (virsh create guest), then udev returns to a
> speedy few seconds.  It seems that the longer the guest
> is up, the longer it takes udev (the +2 hour udev window
> was on a guest up for 6 days) during a guest reboot.
> It's always fast after a destroy/create.
>
> Does anyone have any idea what could be causing the slowness
> in boot and the variability of the time spent in udev?  Any
> suggestions on where we should look?
>
> Thanks,
> Kenton
>
> Specifics:
> ===========
> Host OS:
>     - RHEL 6.1 x64: (2.6.32-131.6.1.el6.x86_64)
>     - qemu-kvm: qemu-kvm-0.12.1.2-2.160.el6_1.2.x86_64.rpm
>     - libvirt: libvirt-0.8.7-18.el6.x86_64.rpm
>
> Guest OS:
>     - RHEL 5.6: (2.6.18-238.19.1.el5)
>
> VM details:
>     - 1 VM/guest on the host
>     - VM size is 32GB ram and 10 cores (host has
>         36GB and 12 cores total)
>         - virtual disk drive is a local file on a RAID 1 disk
>         - cpu pinning set to force each virtual core to
>         a unique core (hyperthreading is turned on)
>         - virtio for storage and network devices
>     - have 16 ethernet devices tied to 16 bridges
>       mapped to 2 NICS on the host
>     - Use of iPXE and SGA bios for VM.
>
>