This issue is specific to Advanced Server 2.1.
When a device disappears, we normally return DID_NO_CONNECT
with sense data indicating UNIT_ATTENTION and device's
removable bit set. With Redhat 7.2 and generic Linux kernel,
this would allow the application ("dd") to just error out.
With Advanced Server, the scsi layer seems to get into
an infinite loop and panic the system.
When we tried other error codes, most of the time,
the scsi layer would re-send the commands forever until the device comes
back.
What kind of error code should we return?
Regis
_______________________________________________
Redhat-devel-list mailing list
Redhat-devel-list@redhat.com
https://listman.redhat.com/mailman/listinfo/redhat-devel-list
