On 01/28/13 21:59, Mike Christie wrote:
On 01/28/2013 02:55 PM, torn5 wrote:
Mike, thanks for your reply
I actually need the blockdevice to be available (in blocked mode) just
after starting the initiator, even if the initiator wasn't able to
connect yet.
Ah ok. That might not be possible. We have to be able to establish a
tcp/ip connection before the scsi layer and udev can setup a block
device. We have to have that connection so the scsi layer can ask the
target what devices exist.
So if the target is not up to the point where it is accepting
connections then there is nothing we can do.
However, if you had already connected to the target, then brought the
target down, the block devices will stay setup while the target is down.
They will just queue IO or fail IO depending on your settings. Not sure
if that helps.
Would it be possible to export from localhost (initiator host) a fake
version of the devices, so for iscsiadm to connect to localhost and
create the blockdevices, then disconnect from there keeping the
blockdevices available, and then try connecting to the real target?
For this to work I need a kind of "target migration" feature of
open-iscsi: is it possible to reassign the target of the connection to
another IP:port on the fly, maybe by writing parameters to sysfs,
restarting iscsid or the like?
Thank you
--
You received this message because you are subscribed to the Google Groups
"open-iscsi" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/open-iscsi?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.
