Debian – Xen: synchronizing time with Dom0

debianntpxen

I have Xen virtual machines running on Debian 6 (both Dom0 and DomU).
I would like to synchronize clock on DomU with Dom0.
On http://wiki.xen.org/wiki/Xen_FAQ_DomU I have read that I need to set /proc/sys/xen/independent_wallclock to 0 but I do not even have /proc/sys/xen directory on DomU:

# ls /proc/sys/xen
ls: cannot access /proc/sys/xen: No such file or directory

What should I do to synchronize clock on DomU with Dom0?

Update:

[    0.000000] Booting paravirtualized kernel on Xen
[    0.000000] Xen version: 4.0.1 (preserve-AD) (dom0)

Best Answer

If you are running Xen 3 or earlier, set proc/sys/xen/independent_wallclock to 0 and the clocks should synchronize automatically.
If you are arunning Xen 4+, set tsc_mode=2 in the vm.cfg file for the VM.

Synchronizing automatically can cause problems with some software. Your OS should handle it, though it might complain about time going backward. The most notable problems are ones where time jumping backward causes something in the program to misbehave (it's somewhat rare, but happens).

If you can't run automatic time synchronization for any reason, set independent_wallclock to 1 or tsc_mode to 0 and run NTP. This will still have drifting problems as the clock will be inconsistent (it's the nature of VMs and really can't be "fixed").

Related Solutions

Linux – Xen DomU root filesystems becoming read-only on iSCSI virtual IP failover

I eventually solved this by using the following advice and settings from the open-iscsi documentation:

8.2 iSCSI settings for iSCSI root
---------------------------------

When accessing the root parition directly through a iSCSI disk, the
iSCSI timers should be set so that iSCSI layer has several chances to try to
re-establish a session and so that commands are not quickly requeued to
the SCSI layer. Basically you want the opposite of when using dm-multipath.

For this setup, you can turn off iSCSI pings by setting:

node.conn[0].timeo.noop_out_interval = 0
node.conn[0].timeo.noop_out_timeout = 0

And you can turn the replacement_timer to a very long value:

node.session.timeo.replacement_timeout = 86400

After setting up the connection to each LUN as described above, the failover works like a charm, even if it takes several minutes to happen.

Debian Lenny as Xen domU fails to boot

mount: cannot read /etc/fstab: No such file or directory

That's a pretty clear sign that your initramfs is borked. Probably because your install is borked.

How did you get past partitioning in debian-installer? Last time I did it (ummm... yesterday) I had to export something that I could partition, so your disk being hda1 is rather weird. Mine looks like:

disk = ['phy:/dev/GLaDOS/xen-portaltest,xvda,w']

its on a logical volume, not a file, but that shouldn't matter. Giving it xvda1 or whatever didn't work; it wanted to partition that, which is fairly silly.

With xvda, I went ahead and partitioned that (into a xvda1 for /boot and an xvda2 for a LVM physical volume, but you could certainly just use that for root). The installer then completed normally, and it works after dealing with the bootloader not executable error documented on the Debian Wiki's Xen entry.

Best Answer

Related Solutions

Linux – Xen DomU root filesystems becoming read-only on iSCSI virtual IP failover

Debian Lenny as Xen domU fails to boot

Related Topic