>>> On 10/3/2014 at 01:22 PM, Neale Ferguson <[email protected]> wrote: 
> A topic that has already been discussed but may be some more information.
> 
> I have a minidisk 0200 that has been CMS formatted:
> 
> [root@rh7cn1 ~]# cat /proc/dasd/devices
> :
> 0.0.0200(ECKD) at ( 94:    16) is dasde       : active at blocksize: 4096, 
> 90000 blocks, 351 MB
> 
> I note it already appears to have a partition on it!
> 
> [root@rh7cn1 ~]# cat /proc/partitions 
> major minor  #blocks  name
> :
>   94       16     360000 dasde
>   94       17     359988 dasde1

Yes, that's the default behavior for FBA devices, and has been for a very long 
time.

> Just for laughs I try and repartition the device:
> 
> [root@rh7cn1 ~]# fdasd /dev/dasde
> 
> fdasd error:  Unsupported disk format
> /dev/dasde is not formatted with z/OS compatible disk layout!

Which would be expected for an FBA device.

> Strange. Okay, let's do a low-level format:
> 
> [root@rh7cn1 ~]# dasdfmt /dev/dasde
> Please enter the blocksize of the formatting [4096]: 
> Drive Geometry: 500 Cylinders * 15 Heads =  7500 Tracks
> 
> I am going to format the device /dev/dasde in the following way:
>    Device number of device : 0x200
>    Labelling device        : yes
>    Disk label              : VOL1
>    Disk identifier         : 0X0200
>    Extent start (trk no)   : 0
>    Extent end (trk no)     : 7499
>    Compatible Disk Layout  : yes
>    Blocksize               : 4096
> 
> --->> ATTENTION! <<---
> All data of that device will be lost.
> Type "yes" to continue, no will leave the disk untouched: yes
> Formatting the device. This may take a while (get yourself a coffee).
> dasdfmt: (invalidate first track) IOCTL BIODASDFMT failed. (Input/output 
> error)

Now that looks like a bug.

> Yech! Here's what /var/log/messages has to say -
> 
> [root@rh7cn1 ~]# tail -20 /var/log/messages
> :
> Oct  3 13:04:48 rh7cn1 kernel: dasd-eckd 0.0.0200: An error occurred in the 
> DASD device driver, reason=09
> Oct  3 13:04:48 rh7cn1 kernel: dasd(eckd): I/O status report for device 
> 0.0.0200:
> dasd(eckd): in req: 000000002d5c3e90 CC:01 FC:04 AC:00 SC:17 DS:02 CS:00 
> RC:0
> dasd(eckd): device 0.0.0200: Failing CCW: 000000002d5c3f98
> dasd(eckd): Sense(hex)  0- 7: 80 00 00 00 00 00 00 01
> dasd(eckd): Sense(hex)  8-15: 00 00 00 00 00 00 00 00
> dasd(eckd): Sense(hex) 16-23: 00 00 00 00 00 00 00 00
> dasd(eckd): Sense(hex) 24-31: 00 00 40 80 00 00 00 00
> dasd(eckd): 24 Byte: 0 MSG 1, no MSGb to SYSOP
> Oct  3 13:04:48 rh7cn1 kernel: dasd(eckd): Related CP in req: 
> 000000002d5c3e90
> dasd(eckd): CCW 000000002d5c3f90: E7400040 2D5C3FA8 DAT:  00a00000 00000000  
> 00000000 00c40000  00000000 00000000  00000000 00000000
> dasd(eckd): CCW 000000002d5c3f98: 47400010 2D5C3FE8 DAT:  03800001 00000000  
> 00000000 00000008
> dasd(eckd): CCW 000000002d5c3fa0: 1D200008 2D5C3FF8 DAT:  00000000 01000000
> Oct  3 13:04:48 rh7cn1 systemd-udevd: inotify_add_watch(7, /dev/dasde1, 10) 
> failed: No such file or directory
> 
> Command reject on a locate record??!! I do note that there was an APAR for 
> z/VM 5.3 that dealt with the x'e7' CCW and command reject on guest virtual 
> machines, but I dare say that's been in z/VM 6.2 from the get-go.  There are 
> no messages on the OPERATOR console indicating z/VM has detected a real I/O 
> error. The only other twist is that I'm running on a zPDT. I thought a prefix 
> CCW contained the define-extend and locate-record information making the 
> x'47' 
> CCW redundant, but I'm no expert with these new fangled prefix thingies.
> 
> Other details: 
> 1. Kernel: 3.10.0-123.8.1.el7.s390x
> 2. s390utils: 1.23.0-12
> 3. z/VM: 6.2 1101

Mark Post

----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
----------------------------------------------------------------------
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Reply via email to