[TriLUG] Xen HVM guest has severe clock drift

[Igor Partola]

Good morning.

I recently rented a VPS from a provider that runs on a Xen HVM
platform. I am running Ubuntu 10.10 i686 (2.6.35-22-generic-pae) on
it. I continuously run ntpd, but the clock drifts by as much as 30
seconds in 5 minutes and NTP cannot keep up. Has anyone experienced
this? What could be the root cause of this problem?

Here are some details:

$ dmesg | grep clock
[    0.160000] Measured 347 cycles TSC warp between CPUs, turning off TSC clock.
[    0.396000] * this clock source is slow. Consider trying other clock sources
[    0.550448] Switching to clocksource acpi_pm
[    0.653135] rtc_cmos 00:05: setting system clock to 2011-03-09
02:45:40 UTC (1299638740)

$ cat /sys/devices/system/clocksource/clocksource0/available_clocksource
acpi_pm

$ cat /sys/devices/system/clocksource/clocksource0/current_clocksource
acpi_pm

So far I've tried the following things:

1. I tried disabling ACPI.
2. I replaced openntpd with ntp and added "tinker panic 0" to the top
of my /etc/ntp.conf file
3. I read up about xen.independent_wallclock flag, which from what I
can tell applies only to Xen PV.
4. Contacted the hosting provider who had little useful advice, aside
from "you should install an ntp deamon."

What could be the root cause of this problem? Is there any sort of
remedy aside from just running an ntp daemon and losing lots of clock
time?