On 11/15/24 12:45, "Kai Mäkisara (Kolumbus)" wrote:
The device->was_reset flag stays set in many (most?) cases forever, unless someone resets it. (Scsi_error resets it only if the associated kthread ends up locking the device door.) Because the flag stays set, the st device can notice it even if the sg device gets the UA.
Yes, this is my observation too. Once was_reset is set, it stays on until a mt load or mt unload command is done. Rewinding or repositioning the tape does not clear it. Perhaps this is what the code in st.c was there to clear was_reset.
Using a flag set by an other layer is ugly (to put it mildly). Even uglier is that st clears the flag when it has noticed that it is set. And there are cases where the device reset does not originate from the same computer.
Yes, this is the original case that caused the problem. A true third party device reset comes from another host, connected to the same SCSI tape lun, through a different IT nexus.
/John
