After switching from iscsitgt to COMSTAR, for whatever reason
pkg image-update now fails to complete more or less at the same
point every time when all the mpt disks on the target suddenly go
offline at the same moment, necessitating a reboot of the target.
This has nothing to do with pkg, but it does leave the install on the
iscsi initiator (a VirtualBox VM) in an odd state.

The install is terminating during the install phase when the new
BE has already been created but not activated, with 30626/43548
whatevers done. I wonder what is special about #30267 :).

Trying to simply restart the update fails with a python dump that
says the be-name already exists, which is true. If I delete the BE,
it will start the install from scratch, which is a waste of resources
and time. Is there a way to continue the update or do I really have
to start all over again? If so, I guess we'll go back to iscsitgt because
COMSTAR seems to be very fragile, to put it mildly, and this never
happened with iscsitgt during many a pkg image-update.

Just trying to go from b125 to b132 on an x86 VM with an iscsi disk.

Thanks



_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss

Reply via email to