mfii0 no disks

[Dima Veselov <kab00m%lich.phys.spbu.ru@localhost>]

Greetings,

I have an IBM x3200 M3 server with IBM M5110 RAID controller. NetBSD 9.3
was installed and was booting, working, etc. I had to replace disks,
made new virtual drive (VD) and removed old one, I did it on temporary linux
using StorCLI utility. After reboot NetBSD refuse to detect disks on
scsibus0. scsictl scsibus0 reset calls inappropriate ioctl, scan all all
does nothing.

[root@ ~]$ dmesg|grep mfii
[     1.091211] mfii0 at pci2 dev 0 function 0: "ServeRAID M5110", firmware 23.34.0-0023, 512MB cache
[     1.091211] mfii0: interrupting at ioapic0 pin 16
[     1.091211] scsibus0 at mfii0: 64 targets, 8 luns per target
[     1.423119] mfii0: physical disk inserted id 9 enclosure 252
[     1.423119] mfii0: physical disk inserted id 12 enclosure 252

[root@ ~]$ dmesg|grep scsi
[     1.091211] scsibus0 at mfii0: 64 targets, 8 luns per target
[     1.398105] scsibus0: waiting 2 seconds for devices to settle...

[root@ ~]$ bioctl mfii0 show
Volume Status       Size         Device/Label    Level Stripe
=============================================================
     0 Online       2.7T                        RAID 1    N/A  65535 seconds
   0:0 Online       2.7T        1:10.0 noencl <ATA     WDC WD30EFRX-68E0A80>
   0:1 Online       3.6T         1:9.0 noencl <ATA     ST4000NM0115-1YZSN04>

It seems that new DG have SCSI target number changed and now it is not 
starting from 0 what possibly can confuse driver.

root@grml ~ # lsscsi -g
[0:2:1:0]    disk    IBM      ServeRAID M5110  3.46  /dev/sda   /dev/sg0 

Is there any kernel logic preventing targets if there are no zero target?

--
Sincerely yours,
Dima Veselov
Physics R&D Establishment of Saint-Petersburg University