Date: Wed, 12 Oct 2016 18:10:21 +0000 From: bugzilla-noreply@freebsd.org To: freebsd-bugs@FreeBSD.org Subject: [Bug 212914] CAM: SATA drives are getting deleted and then re-added after controller rescan Message-ID: <bug-212914-8-XmJrqbm2fR@https.bugs.freebsd.org/bugzilla/> In-Reply-To: <bug-212914-8@https.bugs.freebsd.org/bugzilla/> References: <bug-212914-8@https.bugs.freebsd.org/bugzilla/>
next in thread | previous in thread | raw e-mail | index | archive | help
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=212914 --- Comment #2 from Kenneth D. Merry <ken@FreeBSD.org> --- The device may be going away for several possible reasons: 1. A CCB is returned with the status CAM_DEV_NOT_THERE or CAM_SEL_TIMEOUT. 2. A CCB is returned with the SCSI ASC/ASCQ 0x25,0x00, Logical Unit Not Supported. 3. Someone is doing an xpt_async(AC_LOST_DEVICE, ...) A device may go away and come back as a result of a rescan if any of the following changes: SCSI Standard Inquiry Data (Full inquiry data, including Vendor, Product, Revision) SCSI page 0x80 serial number The first thing I would look at here is what status is getting returned from the drive in question after a reset. If that all looks good, look at whether someone is issuing a rescan, and whether the device is returning inconsistent results. Those inconsistent results could be buried in the part of the Inquiry data that isn't displayed. Standard inquiry data is checksummed along with the serial number and any change in the checksum will make the device go away and come back. Although this is a SATA drive, obviously the only thing that matters is the SCSI response, because CAM is communicating with it as if it is a SCSI protocol device. -- You are receiving this mail because: You are the assignee for the bug.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-212914-8-XmJrqbm2fR>