auto-boot

Fo the LAN card, if you do have the right driver, it should be seen after a
simple "boot -r". Sometimes, one's must add some packages (i.e. for X25
cards).
To be check that the hardware see this device, follow these steps :
# init 0  -> goes to obp (open boot prom)
and gives you this prompt :
ok
enter the following :
ok setenv auto-boot? false
ok reset  (the wait a while)
ok watch-net-all

Then your LAN card should be seen... If not, you might need a driver on
Solaris
to handle it...

then, to get everything as before :
ok setenv auto-boot? true
ok boot -r

As far as I know, there's no way to make a tape bootable on Solaris. May be
on
next release (feb 2002).


Back to the Index

Sun Hardware Diagnostics

Home: www.packetnexus.com

This is a multi-part message in MIME format. ------_=_NextPart_001_01C1A091.A42C6240 Content-Type: multipart/alternative; boundary="----_=_NextPart_002_01C1A091.A42C6240" ------_=_NextPart_002_01C1A091.A42C6240 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Sun Hardware Diagnostics=20 =20 The POST-based hardware diagnostics only check out the devices and buses required to access I/O devices; they do not check the devices themselves. Even so, the onboard hardware diagnostics can often pinpoint the source of a hardware failure.=20 To run Sun hardware diagnostics, perform the following at the ok> prompt:=20 ok> setenv auto-boot? false ok> setenv diag-switch? true ok> setenv diag-level max ok> setenv diag-device disk net (if appropriate) ok> reset (watch results of diagnostic tests)=20 If devices appear to be missing, you can also run the following tests: ok> probe-scsi-all ok> probe-sbus ok> show-sbus ok> show-disks ok> show-tapes ok> show-nets ok> show-devs=20 In addition, the following commands can be used to examine the CPUs or switch to another CPU: ok> module-info ok> processor_number switch-cpu=20 Sometimes additional information can be obtained by navigating the PROM device tree <promnav.html>. You can also look at the OBP manual <http://sundocs.princeton.edu:8888/ab2/coll.216.1/@Ab2CollView?> for additional information on PROM monitor diagnostics.=20 At the end of this process, reset your PROM parameters: ok> setenv auto-boot? true (if appropriate) ok> setenv diag-switch? false (if appropriate)=20 (Note that the diagnostics can take a substantial amount of time to run, depending on your hardware configuration. Most admins prefer to turn them off unless they are diagnosing a problem.)=20 For sun4u (Ultra) systems, you can get some of this information by running /usr/platform/`arch -k`/sbin/prtdiag -v on a running system.=20 Results from the above should be compared to log entries in /var/adm/messages or console error messages.=20 Additional PROM Diagnostics=20 Some additional PROM diagnostics are available at the ok> prompt. To discover what additional diagnostics are available for your hardware, type help diag at the ok> prompt. The output will include the appropriate syntax for all available PROM diagnostic functions. Note that reset should be run as above before running the tests. It is also possible that test-all might hang the system, requiring a power cycle.=20 ------_=_NextPart_002_01C1A091.A42C6240 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN"> <HTML> <HEAD> <META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; = charset=3Diso-8859-1"> <META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version = 6.0.4712.0"> <TITLE>Solaris#Sun Hardware Diagnostics</TITLE> </HEAD> <BODY>  Sun Hardware = Diagnostics <IMG SRC=3D"No%20AttachName" alt=3D"ole0.bmp"> The POST-based hardware diagnostics = only check out the devices and buses required to access I/O devices; = they do n

Sun Hardware Diagnostics

Home: www.packetnexus.com

Sun Hardware Diagnostics


The POST-based hardware diagnostics only check out the devices and buses
required to access I/O devices; they do not check the devices themselves.
Even so, the onboard hardware diagnostics can often pinpoint the source of a
hardware failure.
To run Sun hardware diagnostics, perform the following at the ok> prompt:
ok> setenv auto-boot? false
ok> setenv diag-switch? true
ok> setenv diag-level max
ok> setenv diag-device disk net (if appropriate)
ok> reset
(watch results of diagnostic tests)
If devices appear to be missing, you can also run the following tests:
ok> probe-scsi-all
ok> probe-sbus
ok> show-sbus
ok> show-disks
ok> show-tapes
ok> show-nets
ok> show-devs
In addition, the following commands can be used to examine the CPUs or
switch to another CPU:
ok> module-info
ok> processor_number switch-cpu
Sometimes additional information can be obtained by navigating the PROM
device tree <promnav.html>. You can also look at the OBP manual
<http://sundocs.princeton.edu:8888/ab2/coll.216.1/@Ab2CollView?> for
additional information on PROM monitor diagnostics.
At the end of this process, reset your PROM parameters:
ok> setenv auto-boot? true (if appropriate)
ok> setenv diag-switch? false (if appropriate)
(Note that the diagnostics can take a substantial amount of time to run,
depending on your hardware configuration. Most admins prefer to turn them
off unless they are diagnosing a problem.)
For sun4u (Ultra) systems, you can get some of this information by running
/usr/platform/`arch -k`/sbin/prtdiag -v on a running system.
Results from the above should be compared to log entries in
/var/adm/messages or console error messages.
Additional PROM Diagnostics
Some additional PROM diagnostics are available at the ok> prompt. To
discover what additional diagnostics are available for your hardware, type
help diag at the ok> prompt. The output will include the appropriate syntax
for all available PROM diagnostic functions. Note that reset should be run
as above before running the tests. It is also possible that test-all might
hang the system, requiring a power cycle.


Back to the Index

StorEDGE D1000 with DiskSuite RAID 5

Home: www.packetnexus.com

 -----Original Message-----
From: 	Jason Lewis  
Sent:	Monday, March 04, 2002 4:19 PM
To:	Ga (E-mail)
Subject:	Solaris#StorEDGE D1000 with DiskSuite RAID 5

StorEDGE D1000 with DiskSuite RAID 5

Here's my original question:
*****
Hi. I desparately need help getting a StorEDGE D1000
up and working as a single RAID-5 disk. I need to get
the Ultra10 to recognize the device, need to format the
disks, learn to use DiskSuite, everything.
Right now I'm at the point where I've physically
attached the StorEDGE to the Ultra10, and I've installed
the DiskSuite 4.1 packages, SUNWmd and SUNWmdg.
This is very time-critical, and I would very much appreciate
any assistance. I'm just at the point where I'm trying to
get the system to recognize the StorEDGE device.
I followed one of the manuals, which said to do:
#drvconfig
#disks
then look for device files this way:
#ls -l /dev/dsk/c1t4d0s*
but I'm not getting back anything.
I'm new to adding a disk, so help on even that would be
very much appreciated.
*****
...and here's what I did to set up the RAID:

I brought the Ultra10 down to the OK prompt by using:
# init 0
then I searched for devices using:
# show-disks
and saw that there were several devices aside from the system disk.
For reasons beyond my current understanding, the use of
# probe-scsi-all
causes a message, "bus error" to repeatedly echo to the standard output,
due to the pci card on my Ultra10.
So, using show-disks worked just as well to make sure that
the cables were set up properly and the Ultra10 was "seeing" the D1000.
I then booted the machine, with the "reconfigure" option:
# boot -r
Once the machine was up and I was logged in as root, I used the format
utility to list off all of the disks.
There were 9 disks listed, 0-8. 0 is the system disk, c0t0d0.
For each of the other 8 disks, I ran the format command
(e.g. format->c1t0d0->format, for the first disk on controller one)
to clear the contents of the disk. Then, I ran the partition command
for each of the eight disks, and made the following partitions:
0	0-7499		16.85GB		(7500/0/0)
...
7	7500-7505	13.80MB		(6/0/0)

The slice 7 partition is for the metadb.
I then ran format->label for each of the newly partitioned disks.

I then ran the metadb command:
(make sure that you've installed DiskSuite and that
/usr/opt/SUNWmd/sbin is in your path)
# metadb -f -a -c1 /dev/dsk/c1t0d0s7
# metadb -a -c1 /dev/dsk/c1t1d0s7
# metadb -a -c1 /dev/dsk/c1...d0s7
(one for each of the eight disks on which I allocated a slice 7)
then I ran a test version of the command to make the RAID system:
# metainit -n d1 -r /dev/dsk/c1t0d0s0 /dev/dsk... -i 2k
(one /dev/dsk entry for each of the disks on which I allocated
a slice 0 for use in the RAID system)
After seeing that the value returned to standard output by this
command was not an error message, I ran the real command:
# metainit d1 -r /dev/dsk/c1t0d0s0 /dev/dsk... -i 2k

Following this invocation of metainit, I followed the progress
of the RAID build using:
# metastat

Once the build was complete, I had an entry,
/dev/md/dsk/d1
which was recognized as a RAID system.

I'm still in the process of setting this up, but that's
fundamentally what I did to get the RAID up.

Thanks a lot to everyone who replied.

Mike Phelan

Back to the Index

Script for Disk Mirroring

Home: www.packetnexus.com

Here is the appropriate script:

#!/bin/sh

ODSINIT=/etc/rc2.d/S78disksuite
METABIN=/usr/opt/SUNWmd/sbin
DISK1=c0t0d0
DISK2=c0t2d0

echo ""
echo "###Creating the file $ODSINIT for configuring ODS###"
echo ""

echo "#!/sbin/sh

if test -f /config.ods.1
then

     $METABIN/metadb -a -f -c 3 ${DISK1}s7 ${DISK2}s7

     echo \"Preparing the / partition\"
     $METABIN/metainit -f d10 1 1 ${DISK1}s0
     $METABIN/metainit d20 1 1 ${DISK2}s0
     $METABIN/metainit d0 -m d10
     $METABIN/metaroot d0

     echo \"Preparing the swap partition\"
     $METABIN/metainit -f d11 1 1 ${DISK1}s1
     $METABIN/metainit d21 1 1 ${DISK2}s1
     $METABIN/metainit d1 -m d11

     echo \"Preparing the /var partition\"
     $METABIN/metainit -f d12 1 1 ${DISK1}s3
     $METABIN/metainit d22 1 1 ${DISK2}s3
     $METABIN/metainit d2 -m d12

     echo \"Preparing /etc/vfstab. Backup in /etc/vfstab.orig\"

     cp /etc/vfstab /etc/vfstab.orig

     echo \"#device         device          mount           FS      fsck 
    mount   mount
#to mount       to fsck         point           type    pass    at boot 
options
#
#/dev/dsk/c1d0s2 /dev/rdsk/c1d0s2 /usr          ufs     1       yes     -
fd      -       /dev/fd fd      -       no      -
/proc   -       /proc   proc    -       no      -
/dev/md/dsk/d1       -       -       swap    -       no      -
/dev/md/dsk/d0       /dev/md/rdsk/d0      /       ufs     1       no 
   logging
/dev/md/dsk/d2       /dev/md/rdsk/d2      /var    ufs     1       no 
   logging,nosuid
swap    -       /tmp    tmpfs   -       yes     nosuid
\" > /etc/vfstab.meta

     lockfs -fa

     if [ -r /dev/md/dsk/d0 ]; then
         cp /etc/vfstab.meta /etc/vfstab
     fi

     touch /config.ods.2
     rm /config.ods.1
     reboot
fi

if test -f /config.ods.2
then
     rm /config.ods.1
     reboot
fi

if test -f /config.ods.2
then
     echo\"\"
     echo \"Meta Attaching ODS devices...\"
     echo\"\"
     $METABIN/metattach d0 d20
     $METABIN/metattach d1 d21
     $METABIN/metattach d2 d22
     rm /config.ods.2
fi
" > /a/$ODSINIT


Back to the Index

ok prompt not booting

Home: www.packetnexus.com

Problem:
We recently recieved a SUN E250 Server with Solaris 2.6 installed. I need
to know the location which I need to edit so the Server does not stop at
the OK prompt and will continue to boot when you power on. Currently, I
need to type boot. When running init 6 it shuts down and comes all the way
up per the init6 start up script. I have searched various startup scripts
and did some research to no avail.

The solution is:
As root, run 'eeprom' and check for:

auto-boot?=true

to correct, you can make the change either in the open boot or at the
command line.

root#  eeprom auto-boot?=true


Back to the Index