> Hehehe. So I was right about the problem was in storage :D
I didn't say it wasn't storage, I thought you asked if it was network storage
> How did you find out it was corrupted? fsck -f SHOULD be able to
> identify SOME problems, which is why I asked to fsck and mount domU's
> disk earlier.
Well, here's the weirdness. Trying to console into the guest got me nowhere and
all I could see is what I posted. So I decided to go into init mode 5, desktop,
then looked at the guest as it was booting and saw that it hung on a disk
problem, requiring fsck. Fine, but, what's weird to me is that rebooting the
host, nothing was wrong with the local storage, it looked just fine but the
guest's storage, which is on the host, was messed up?
I realize that the guest is in it's own world for the most part but I didn't
> If both domU and dom0 located on the same drive, yet only domU is
> corrupted, there is a possibility that somehow you may have mounted it
> twice at the sametime: on dom0 and domU, or on two domUs. Xen (on
> linux anyway) should prevent this, but if you force it to (like
> putting phy://...,w! instead of just w) or if you're on a shared-disk
> setup (two dom0s accessing the same shared storage) it can happen.
Don't recall doing anything special but, there is something I did do with this
particular guest. As I've been setting up hardware, I did try a few manual
commands to see if I could migrate it to another machine. I wonder if I messed
it up when I was doing that.
Interesting but I sure wish there was a way of getting more info from command
line when something goes wrong. Every xm and virsh information test I looked up
returned nothing wrong with the guest, yet there it was, but it took desktop to
I've not found any info on where you automate turning on guests when the host
boots and what I've done so far is to have virsh start in the rc.local. Bet
that might be a problem if I accidentally have guests being turned on somewhere
else when the host boots.
Xen-users mailing list