Hi,
we got exactly same issue with two Dell PowerEdge R900. After running
without any trouble for 50 resp. 60 days, Dom0 does jump several hours
in a future.
Dom0 mean in "xm debug-key s; xm dmesg"
(XEN) Platform timer appears to have unexpectedly wrapped 10 or more times.
(XEN) TSC has constant rate, deep Cstates possible, so not reliable,
warp=15048 (count=1)
(XEN) dom1: mode=0,ofs=0x175a91cd54,khz=2925946,inc=1,vtsc count:
164505657108 kernel, 22209746 user
(XEN) dom3: mode=0,ofs=0xb0d3e99f1f81,khz=2925946,inc=1,vtsc count:
30339779336 kernel, 19604648 user
(XEN) dom4: mode=0,ofs=0xdc739842b4f4,khz=2925946,inc=1,vtsc count:
5903404261 kernel, 29402760 user
(XEN) dom12: mode=0,ofs=0x12f1b269dbff1d,khz=2925946,inc=1,vtsc count:
6914273 kernel, 78450 user
DomU on same Host:
kernel: : [5100661.449288] Clocksource tsc unstable (delta =
-807453857134 ns)
We running hypervisor with boot line "/boot/xen-4.0-amd64.gz placeholder
clocksource=pit cpuidle=0 dom0_mem=512M loglvl=all guest_loglvl=all"
As CPU we have 4x "X7350 @ 2.93GHz"
Any ideas how we can fix or avoid these jumps? Ist it an option to
ignore Clock jumps in Dom0 in DomU (xen.independent_wallclock=1 was in
Xen3 an Option).
We have no problem if Dom0 jumps, but for DomU it is catastrophically.
Regards,
Dimitrij
--
Dimitrij Hilt
dimitrij.h...@fhe3.com
http://www.fhe3.com/
--
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/4f4209dc.1060...@fhe3.com
