On Mon, Mar 19, 2007 at 02:51:57PM +1200, Nicholas Lee wrote: > bash-3.00# iscsitadm list target > iscsitadm: Bad XML response > bash-3.00# iscsitadm list target -v > iscsitadm: Bad XML response > > Any suggestions why this is happening?
This can happen if the daemon isn't correctly listening for requests either because it's still starting up, because it's dumping core or for some other bug. Did it produce a core file? If you haven't already, I strongly suggest you configure coreadm(1M) to enable the global core file repository and then check to see if you get any core files from the iSCSI target daemon. > When I'm only using zfs shareiscsi I need to restart the target service or > destroy the volume and recreate for the xml config file to be recreated. This sounds like a bug in the iSCSI target. There have been several fixes in this area recently -- what build are you running? > I'm not sure where zfs stores its iscsi information, but when I also use the > manual (iscsitadm create target -b ...) method the config.xml for this seems > to stay in place. ZFS stores its iSCSI configuration information as a property of the zvol dataset (that is, not in the xml file in /etc). > Regardless I'm having issues with trying to connect to the iscsi target. Its > broken some how. I tried a quick test from an openfiler vmware image I've > got some PDU errors trying to start the initiator against the b55b Solaris > target. Any details you can provide would be very helpful. If you want to file a bug directly with that information, that would be even more helpful. > I'm surprised it's such a pita, makes me wonder if zfs is really worth the > pain of solaris vs linux. Sorry you're finding the transition a bit rocky. We've been working to improve the iSCSI target, ZFS, and their interactions. Adam -- Adam Leventhal, Solaris Kernel Development http://blogs.sun.com/ahl _______________________________________________ storage-discuss mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/storage-discuss
