Thank you Mark for you response. I have my responses inline.
Mark Johnson wrote:
Should only take 20 minutes or so.
I knew I wasn't doing it right.
How is your dom0 configured?
Basically enabled xVM on the system.
How many CPUs in the system?
The system is a Sun Fire X4150 with 1 Quad-Core Intel processor.
Are you using zfs?
If so, are you limiting the ARC cache
No.
Are you using a dom0_mem in menu.lst>
No, I haven't used dom0_mem. As I see in
http://www.opensolaris.org/os/community/xen/docs/configuring-dom0/
For a system with 32GB memory I should set dom0_mem to 8GB.
Did you limit ballooning in dom0?
Are you restricting the number of CPUs dom0 uses?
Why aren't you using virt install?
Actually I did use virt install. I only dumped the xml as I thought it
would be easier to read.
Why aren't you using tap vs file?
I wasn't aware of this. Is there a URL where I can read more about such
things besides
http://www.opensolaris.org/os/community/xen/docs/configuring-dom0/
In the configuratin below I would set driver name=tap instead?
This is problem # 1... Your using file instead of
tap, and it's probably on a zfs based system.
lofi runs *really* slow on top of zfs.
<disk type='file' device='disk'>
<driver name='file'/>
<source file='/xvm/lm_ufs.img'/>
<target dev='xvda'/>
MRJ
Here is what I had defined in my xml :
<domain type='xen' id='-1'>
<name>snv115_nfs_12gb</name>
<uuid>e3c1a3c2-0bda-65ab-3c7e-8c6acf180dd0</uuid>
<bootloader>/usr/lib/xen/bin/pygrub</bootloader>
<os>
<type>solaris</type>
</os>
<memory>12582912</memory>
<vcpu>1</vcpu>
<on_poweroff>destroy</on_poweroff>
<on_reboot>restart</on_reboot>
<on_crash>restart</on_crash>
<distro name='solaris'/>
<clock offset='localtime'/>
<devices>
<interface type='bridge'>
<source bridge='e1000g0 '/>
<target dev='vif-1.0'/>
<mac address='00:16:3e:45:94:ef'/>
</interface>
<disk type='file' device='disk'>
<driver name='file'/>
<source file='/xvm/lm_ufs.img'/>
<target dev='xvda'/>
</disk>
</devices>
</domain>
thanks,
Sagun
_______________________________________________
xen-discuss mailing list
[email protected]
_______________________________________________
xen-discuss mailing list
[email protected]
