[Gluster-devel] i can't run a xen - virtual machine on a glusterfs mount point
Jordi Moles Blanco
jordi at cdmon.com
Mon Jul 7 16:48:52 UTC 2008
hi,
i've been using glusterfs for a while now in different scenarios and now
i'm facing a really strange problem, or at least it seems so to me....
the thing is that i've got a glusterfs, with six nodes and several
clients (xen-servers) mounting the shared system. The idea is to have a
big shared storage for all the vm's i've got.
when i try to run a vm created from xen 3.2 which is stored on gluster,
it won't run.
it says:
Error: Device 2049 (vbd) could not be connected. Backend device not found.
nodes don't log any error or warning at all, and the xen-server which is
trying to run the vm says this:
2008-07-07 17:43:07 E [afr.c:2391:afr_writev_cbk] grup1:
(path=/domains/xen_testdedicat1/disk.img child=espai2) op_ret=-1 op_errno=22
2008-07-07 17:43:07 E [fuse-bridge.c:1645:fuse_writev_cbk]
glusterfs-fuse: 656: WRITE => -1 (22)
2008-07-07 17:43:07 E [afr.c:2699:afr_flush] grup1: afrfdp->fdstate[] is
0, returning ENOTCONN
2008-07-07 17:43:07 E [fuse-bridge.c:945:fuse_err_cbk] glusterfs-fuse:
657: (16) ERR => -1 (107)
i'm using this software versions
linux debian etch, kernel 2.6.18-6-xen-amd64
fuse-2.7.3glfs10
glusterfs--mainline--2.5--patch-788
the thing is, however, that if i move the virtual machine create from
xen 3.2 from the shared gluster mount point to a local path, it will run
without any problem at all...
is it that gluster is not fast enough to provide xen with the data
needed to launch the vm? could i do anything to fix that?
these are the specs files:
node side:
***********
volume espa
type storage/posix
option directory /mnt/compartit
end-volume
volume spai
type performance/io-threads
option thread-count 4
option cache-size 32MB
subvolumes espa
end-volume
volume espai
type performance/write-behind
option aggregate-size 1MB
option flush-behind on
subvolumes spai
end-volume
volume ultim
type protocol/server
subvolumes espai
option transport-type tcp/server
option auth.ip.espai.allow *
end-volume
***********
client/xen side:
***********
volume espai1
type protocol/client
option transport-type tcp/client
option remote-host 192.168.1.204
option remote-subvolume espai
end-volume
volume namespace1
type protocol/client
option transport-type tcp/client
option remote-host 192.168.1.204
option remote-subvolume nm
end-volume
volume espai2
type protocol/client
option transport-type tcp/client
option remote-host 192.168.1.206
option remote-subvolume espai
end-volume
volume espai3
type protocol/client
option transport-type tcp/client
option remote-host 192.168.1.213
option remote-subvolume espai
end-volume
volume espai4
type protocol/client
option transport-type tcp/client
option remote-host 192.168.1.161
option remote-subvolume espai
end-volume
volume namespace2
type protocol/client
option transport-type tcp/client
option remote-host 192.168.1.161
option remote-subvolume nm
end-volume
volume espai5
type protocol/client
option transport-type tcp/client
option remote-host 192.168.1.162
option remote-subvolume espai
end-volume
volume espai6
type protocol/client
option transport-type tcp/client
option remote-host 192.168.1.163
option remote-subvolume espai
end-volume
volume grup1
type cluster/afr
subvolumes espai1 espai2
end-volume
volume grup2
type cluster/afr
subvolumes espai3 espai4
end-volume
volume grup3
type cluster/afr
subvolumes espai5 espai6
end-volume
volume nm1
type cluster/afr
subvolumes namespace1 namespace2
end-volume
volume ultim
type cluster/unify
subvolumes grup1 grup2 grup3
option scheduler rr
option namespace nm1
end-volume
volume iot
type performance/io-threads
option thread-count 4
option cache-size 32MB
subvolumes ultim
end-volume
volume ioc
type performance/io-cache
option cache-size 64MB
option page-size 1MB
option force-revalidate-timeout 2
subvolumes iot
end-volume
***********
i've already tried for a while to play with the spec file... adding and
removing write and read buffers... but the final result is the same...
the vm won't run.
any idea?
thank you.
More information about the Gluster-devel
mailing list