ia64 panic: make_dev_credv: bad si_name (error=17, si_name=pass3)

[Anton Shterenlikht]

	From jhb at freebsd.org Thu Sep 20 20:30:06 2012

	On Thursday, September 20, 2012 4:53:30 am Anton Shterenlikht wrote:
	> On ia64 r235474 I added another
	> disk and did "camcontrol rescan all"
	> to see it. I got this panic:
	> 
	> panic: make_dev_credv: bad si_name (error=17, si_name=pass3)


	17 is EEXIST.  It seems you still had a pass3 device present (that is the
	root cause of this panic).  Presumably you pulled a disk first and that didn't
	get cleaned up properly?  Either that or CAM is trying to use the same name
	for two devices for some reason.

Ok, what I did exactly was this.

This node had 3 scsi disks directly inside it:

da0 at mpt0 bus 0 scbus0 target 0 lun 0
da0: <HP 73.4G ST373454LC HPC2> Fixed Direct Access SCSI-3 device 
da0: 320.000MB/s transfers (160.000MHz, offset 63, 16bit)
da0: Command Queueing enabled
da0: 70007MB (143374738 512 byte sectors: 255H 63S/T 8924C)
da1 at mpt0 bus 0 scbus0 target 1 lun 0
da1: <SEAGATE ST318452LC 2213> Fixed Direct Access SCSI-3 device 
da1: 160.000MB/s transfers (80.000MHz, offset 63, 16bit)
da1: Command Queueing enabled
da1: 17366MB (35566478 512 byte sectors: 255H 63S/T 2213C)
da2 at mpt1 bus 0 scbus1 target 2 lun 0
da2: <SEAGATE ST318452LC 2213> Fixed Direct Access SCSI-3 device 
da2: 160.000MB/s transfers (80.000MHz, offset 63, 16bit)
da2: Command Queueing enabled
da2: 17366MB (35566478 512 byte sectors: 255H 63S/T 2213C)

and another 2 logical disks attached via fibre, isp(4):

da3 at isp0 bus 0 scbus2 target 0 lun 1
da3: <COMPAQ MSA1000 VOLUME 4.32> Fixed Direct Access SCSI-4 device 
da3: 200.000MB/s transfers WWNN 0x500805f3000ec220 WWPN 0x500805f3000ec221 PortID 0x10000
da3: Command Queueing enabled
da3: 69460MB (142255575 512 byte sectors: 255H 63S/T 8855C)
da4 at isp0 bus 0 scbus2 target 0 lun 5
da4: <COMPAQ MSA1000 VOLUME 4.32> Fixed Direct Access SCSI-4 device 
da4: 200.000MB/s transfers WWNN 0x500805f3000ec220 WWPN 0x500805f3000ec221 PortID 0x10000
da4: Command Queueing enabled
da4: 140011MB (286744185 512 byte sectors: 255H 63S/T 17849C)

I then cut out another logical disk from MSA1000,
and assigned it lun 2. So if da(4) numbers disks
in the order of their lun, then I guess there was
a conflict - lun 2 disk had to become da4, but
da4 already existed. Is that what you refer to?

Anyway, after the panic and a reboot the former
da4 became da5, and the new disk became da4:

da3 at isp0 bus 0 scbus2 target 0 lun 1
da3: <COMPAQ MSA1000 VOLUME 4.32> Fixed Direct Access SCSI-4 device 
da3: 200.000MB/s transfers WWNN 0x500805f3000ec220 WWPN 0x500805f3000ec221 PortID 0x10000
da3: Command Queueing enabled
da3: 69460MB (142255575 512 byte sectors: 255H 63S/T 8855C)
da4 at isp0 bus 0 scbus2 target 0 lun 2
da4: <COMPAQ MSA1000 VOLUME 4.32> Fixed Direct Access SCSI-4 device 
da4: 200.000MB/s transfers WWNN 0x500805f3000ec220 WWPN 0x500805f3000ec221 PortID 0x10000
da4: Command Queueing enabled
da4: 69460MB (142255575 512 byte sectors: 255H 63S/T 8855C)
da5 at isp0 bus 0 scbus2 target 0 lun 5
da5: <COMPAQ MSA1000 VOLUME 4.32> Fixed Direct Access SCSI-4 device 
da5: 200.000MB/s transfers WWNN 0x500805f3000ec220 WWPN 0x500805f3000ec221 PortID 0x10000
da5: Command Queueing enabled
da5: 140011MB (286744185 512 byte sectors: 255H 63S/T 17849C)

So how could I've done what I wanted to do
without causing a panic? Maybe I should've
unmounted all but the root disk?

Thank you

Anton