[Gluster-users] Gluster 1.3.12 and Xen!

[Enrico Valsecchi]

Dear All,

I'm very new Gluster FS user....
...and apologize for my very poor english!

:-)

I have installed Gluster FS v. 1.3.12 on 2 server with Debian 4.05,
in AFR with client side replication, explained in this page:
http://www.gluster.org/docs/index.php/Setting_up_AFR_on_two_servers_with_client_side_replication
Tested to work!

:-)

Now, my idea:

Use a Xen Hypervisor with image disk data for DomU machines on Gluster 
FS repository

In my Gluster FS client machine (based on Centos 5.2), I have create
a DomU machine with image file stored in local disk 
(/var/lib/xen/images/), Called "Debian4"
O.K., start and finish installation without problems!

Now, I have moved image file from /val/lib/xen/images/ to my GlusterFS 
directory (/mnt/glusterfs/)

O.k., checked that image file, I'm go to modify DomU configuration wit 
new image disk path....

First setting:
disk = [ 'file:/mnt/glusterfs/debian40.img,hda1,w' ]

Start DomU, I retry this error:
xend.err 'Device 768 (vbd) could not be connected. losetup /dev/loop0 
/mnt/glusterfs/debian40.img failed'
Into log explained error is:
domid: 22
qemu: the number of cpus is 1
Watching /local/domain/22/logdirty/next-active
xs_read(): vncpasswd get error. 
/vm/353e360e-bcdd-f237-595e-85013704aad7/vncpasswd.
char device redirected to /dev/pts/4
qemu_map_cache_init nr_buckets = 10000
shared page at pfn 1ffff
buffered io page at pfn 1fffd
xs_read(/vm/353e360e-bcdd-f237-595e-85013704aad7/rtc/timeoffset): read error
I/O request not ready: 0, ptr: 0, port: 0, data: 0, count: 0, size: 0
Triggered log-dirty buffer switch

Second setting:
disk = [ 'tap:aio:/mnt/glusterfs/debian40.img,hda1,w' ]
Machine start and stop

Into log explained error is:
qemu: could not open hard disk image 
'aio:/mnt/glusterfs/debian40-dnsldap.img'
Watching /local/domain/21/logdirty/next-active
xs_read(): vncpasswd get error. 
/vm/353e360e-bcdd-f237-595e-85013704aad7/vncpasswd.
char device redirected to /dev/pts/4
qemu_map_cache_init nr_buckets = 10000
shared page at pfn 1ffff
buffered io page at pfn 1fffd
xs_read(/vm/353e360e-bcdd-f237-595e-85013704aad7/rtc/timeoffset): read error
I/O request not ready: 0, ptr: 0, port: 0, data: 0, count: 0, size: 0
Triggered log-dirty buffer switch

O.k., people have a solution?

Many Thanks,

Enrico Valsecchi