On Tue, May 13, 2025 at 12:29:51PM -0400, Benjamin Marzinski wrote: > Without reading that sense and status information in kernel, the > multipath target can't know if it needs to fail a path and retry the > ioctl down a different path. QEMU can read this information, but it > doesn't know what path the multipath device send the ioctl down. This > patch just gives users a way to check the paths in the active pathgroup > (which all should be able to handle IO) and fail those that can't. > While QEMU is the driver of this, it's completely general functionality.
As just replied to Martin the problem is that this setup fundamentally can't work. Either you pass a SCSI devices through, which should (mostly, there are a few warts) work. Or you want host side multipathing, in which case you must pass through the block device abstraction and not a SCSI one, or at least do a full emulation of the SCSI interfaces instead of pretending to pass it through. The root of all evil here is that dm-multipath tries to pass through SG_IO, which is dangerous for all but the most trivial commands.
