On 03/26/2012 03:32 AM, Rene wrote: > Mike Christie <michaelc@...> writes: >> I think you need to rescan the devices at the scsi layer level (like >> doing a echo 1 > /sys/block/sdX/device/rescan) then run some multipath >> to command, then run some FS and LVM commands if needed. > > Hi, > I'm having a similar problem and stumbled over this thread. For me just > running > "multipath -v0" was enough. My question is: Shouldn't multipathd be able to > do > that on its own? Whats the point of having 2 paths if it doesn't > automatically > recover by itself?
Ask the multipath list not the iscsi list. It's probably due to multipath being geared to handling path failures. The errors we see in Ulrich's log are device failures. We see them on all paths since they all point to the same device. multipath knows how to recover from a path failure, but it does not know enough about the problem to do something about the error we see in this case. It could be made to do so, but for "Illegal Request - Logical block address out of range" we normally need changes in the entire kernel. For example you need to rescan the scsi device manually as it is now. multipath should not have to do that. The scsi layer or some other handler should, so it works for non multipath cases. -- You received this message because you are subscribed to the Google Groups "open-iscsi" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/open-iscsi?hl=en.
