2005-11-29

More things I'll forget

I'm back on-line now, in my new place in Newport and in a new job, than seems at present to involve more hacking C++ TCP and XML code than anything AI. But at least I'm not over stressed and they have fast machines and big screens (those Dell 24" jobbies everyone loves).

I killed my iPod nano with a glass of water. Maybe having a black one full of Cure songs made it want to die. It was asking for it dressed like that. And I didn't get to Skin's gig.

Today children, in my ongoing occasional adventures into headless unix admin, I'm trying to hot-swap scsi drives on a Sun Netra t1, so I can get all of the boxes I have up and running (only one of the three boots into a network connectable state, and the laptop I was using to operate LOM is suffering from old age that gaffer tape won't fix, and tends to switch off if you touch any of the keys, so I'm sshing it to minicom the Sun boxes).

The notes follow:

To find if there is scsi hardware there, you can probe-scsi-all from the LOM console login at run level 0:
bash-2.05# sync;sync;init 0
INIT: New run level: 0
The system is coming down. Please wait.
System services are now being stopped.
Print services already stopped.
Nov 29 22:56:48 unknown syslogd: going down on signal 15
The system is down.
syncing file systems... done
Program terminated

Reset without reboot:
ok setenv auto-boot? false
auto-boot? = false
ok reset-all
Resetting ...


Once the ok prompt is up, find the scsi drive:
ok probe-scsi-all
/pci@1f,0/pci@1,1/scsi@2
Target 0
Unit 0 Disk IBM DDYST1835SUN18G S94A
Target 1
Unit 0 Disk FUJITSU MAN3184M SUN18G 1502


Then reboot:
ok setenv auto-boot? true
auto-boot? = true
ok reset-all
Resetting ...


After logging into Solaris, it's cfgadm to configure the attached disk.

First, list all attachment points:
bash-2.05# cfgadm -al
Ap_Id Type Receptacle Occupant Condition
c0 scsi-bus connected configured unknown
c0::dsk/c0t0d0 disk connected configured unknown
c0::sd1 disk connected configured unknown


The added device is c0::sd1; it's not properly configured, so unconfigure it:
bash-2.05# cfgadm -c unconfigure c0::sd1
bash-2.05# cfgadm -al
Ap_Id Type Receptacle Occupant Condition
c0 scsi-bus connected configured unknown
c0::dsk/c0t0d0 disk connected configured unknown
c0::sd1 disk connected unconfigured unknown


Then configure it:
bash-2.05# cfgadm -c configure c0::sd1
bash-2.05# cfgadm -al
Ap_Id Type Receptacle Occupant Condition
c0 scsi-bus connected configured unknown
c0::dsk/c0t0d0 disk connected configured unknown
c0::dsk/c0t1d0 disk connected configured unknown


Now format can see it:
bash-2.05# format
Searching for disks...done


AVAILABLE DISK SELECTIONS:
0. c0t0d0
/pci@1f,0/pci@1,1/scsi@2/sd@0,0
1. c0t1d0
/pci@1f,0/pci@1,1/scsi@2/sd@1,0


That's enough for now.


TME

2 Comments:

Blogger Jean-Marc said...

THANKS !!!
I had exactly the *same problem*. With a unconfigure/configure, all things went well.
You saved my day !

8:35 AM  
Blogger Jean-Marc said...

Actually I did a "devfsadm -C" after an "unconfigure" on a crashed disk. That's why Solaris was a bit lost when I inserted the new disk. It didn't find the block devices in /dev or /devices for it.

8:38 AM  

Post a Comment

<< Home