Summary: things are changing, but I can't tell if it's progress or not.

Hall, Ken (IDS ECCS) wrote:

I used to do the "add device", but I found it's only necessary if the device isn't in the range specified in zipl.conf at boot time. If it's predefined, you get some messages on the console, and the device "magically" appears.


In my case, it was part of the problem.


The steps you need to do are:

1) Create the minidisk and attach to the machine (link or hcp link).

2) Check /proc/dasd/devices and make sure the disk is there. If it isn't, do the command below, but don't forget to add the device to zipl.conf and run zipl to make it permanent.

3) Make sure there are device nodes in /dev for it.  Several scripts have been 
published that read /proc/dasd/devices and build required nodes.  The base major/minor 
numbers are shown in
/proc/dasd/devices.  These reference the WHOLE DISK.  Nodes also have to be built for 
at least one partition.  For example, if /proc/dasd/devices has:

010b(ECKD) at ( 94: 44) is dasdl : active at blocksize: 4096, 108000 blocks, 421 MB

So you would do:

mknod /dev/dasdl b 94 44
mknod /dev/dasdl1 b 94 45
mknod /dev/dasdl2 b 94 46
mknod /dev/dasdl3 b 94 47


Someone (something?) did this for my system a long time ago, if I grok "ls -l /dev/dasd*" correctly. But I can't tell at first glance what major/minor devices are associated with the nodes.

4) Run dasdfmt.


Since adding the devices to /proc/dasd/devices, the results of this are changing.

--- snip here ---
oracle01:~ # dasdfmt -b 4096 -v -f /dev/dasdf1
Retrieving disk geometry...
Drive Geometry: 3338 Cylinders * 15 Heads =  50070 Tracks

I am going to format the device /dev/dasdf1 in the following way:
  Device number of device : 0x1a2
  Labelling device        : yes
  Disk label              : VOL1
  Disk identifier         : 0X01A2
  Extent start (trk no)   : 0
  Extent end (trk no)     : 50069
  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).
Detaching the device...
Invalidate first track...
dasdfmt: (invalidate first track) IOCTL BIODASDFMT failed. (Invalid
argument)


--- snip here ---


Does dasdfmt get upset if I've already CMS-formatted the minidisk? At
any rate, after doing that, "dasdf" has gone from "active" to "accepted"
after that dasdfmt attempt.

Glancing into the LVM screens in YAST (yes, I'm still using training
wheels, YAST), I see two of the five minidisks I tried to add listing --
the three I've tried to "dasdfmt" aren't listed. It looks like I'm
making things worse, not better. How do I make the first three minidisks
look usable to dasdfmt (or Linux in general)?

Thanks for the patience, one and all.... :-)

Nick

----------------------------------------------------------------------
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

Reply via email to