Summary: I was not able to format the disk or create a backup label. I had
power. I tried powering the system down and disconnecting the disk. I
couldn't get it to respond no matter what I did. I'll replace it.
Thanks to all those who offered suggestions:
Mike Barnett
Tom Vayda
Frank Smith
Jeff Kennedy
Harish
Joel Lee
ganeshan
Matt Reynolds
Responses were (original post at the end):
1. Try shutting the system down (shutdown -i0 -g0 -y) and powering completely
off.
2. Check the cables.
3. Power up the external devices first and then the system.
4. Pray that the controllers on those disks didn't burn down.
I hope you have a backup of those disk partitions and a way of getting
ahold of another disk drive to replace.
Mike Barnett
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
+++++
I know this is obvious but sounds like your hardware is gone.
Try switching cables and SCSI adapters. Try the disk on another machine if
you do not have multiple adapters but I think it may be time to replace the
disk.
Tom Vayda
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
+++++++
Just on the off chance you haven't tried it already, shut down the
computer and power off it and everything connected to it. Then start
turning devices back on, starting with the ends of the SCSI chains
and working your way back to the computer, turning the computer back
on last.
This will work if it is just a problem of the disk inteface coming
back up in some undetermined state, as sometimes will happen on power
glitches and it will never respond to the reset signal on the SCSI bus.
There's also the chance that the interface itself is hosed, or
possibly one of the other disks on the chain. Try removing it and
see if the system mounts the other two.
Frank Smith
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
I know this isn't what most UNIX admins want to hear, since it's
considered a M$ fix, but you should power off the system AND the
external drives for at least a minute. We had a problem with one of our
SSA100's that runs Oracle. It rebooted in the middle of the night and
then wouldn't mount the disks at all. Probe-scsi-all worked for us
too. After about an hour of screwing with it we finally turned it all
off, went to the bathroom, came back and started it all up.
Whallah!!!! Like magic.
Jeff Kennedy
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
+++++
I have some information that might be helpful in your equipment planning.
I have seen equipment failure associated with power hits with the disk
controllers, power supplies, CPUs and sundry computer components giving
up the ghost like this.
I installed a SMART UPS from APC on a computer. I also installed the
PowerChute application that came with the SMART UPS. After letting it run
for a while, I found this log file. I would like to forward it to you
to give you an idea the types of line conditions we see here in Fremont,
California. The log file has been helpful in justification for the
purchase of more UPSs.
If you install the UPS and PowerChute software on just one machine and
let it run for a while, you might be able to justify your own purchases
to the executive staff. Executives love simple explanations.
I hope this helps you.
Here is my log file from PowerChute.
Note that a few days after I installed the UPS, there was a blackout.
Note that there is often no explanation why the UPS switches to battery.
Id Date Time Event
100000 05/21/98 19:32:49 *** PowerChute PLUS Version 5.0.1 Started ***
100200 05/21/98 19:32:59 Communication established
200003 05/28/98 10:51:12 UPS on battery: Blackout 010.4 V
100300 05/28/98 10:51:13 Normal power restored: UPS on line
100403 06/04/98 18:20:03 Self-test at UPS passed
200007 06/11/98 04:27:24 UPS on battery: Large momentary spike 122.8 V
100300 06/11/98 04:27:24 Normal power restored: UPS on line
200006 06/17/98 03:00:06 UPS on battery: Deep momentary sag 088.4 V
100300 06/17/98 03:00:06 Normal power restored: UPS on line
100403 06/18/98 18:19:58 Self-test at UPS passed
200007 06/27/98 06:29:12 UPS on battery: Large momentary spike 120.2 V
100300 06/27/98 06:29:12 Normal power restored: UPS on line
200007 06/27/98 15:02:00 UPS on battery: Large momentary spike 121.5 V
100300 06/27/98 15:02:00 Normal power restored: UPS on line
200000 06/27/98 15:02:25 UPS on battery
100300 06/27/98 15:02:25 Normal power restored: UPS on line
200007 07/01/98 22:24:49 UPS on battery: Large momentary spike 118.9 V
100300 07/01/98 22:24:49 Normal power restored: UPS on line
200000 07/02/98 14:43:06 UPS on battery
100300 07/02/98 14:43:06 Normal power restored: UPS on line
100403 07/02/98 18:19:48 Self-test at UPS passed
200007 07/06/98 07:01:26 UPS on battery: Large momentary spike 122.8 V
100300 07/06/98 07:01:26 Normal power restored: UPS on line
200007 07/10/98 22:21:06 UPS on battery: Large momentary spike 119.6 V
100300 07/10/98 22:21:07 Normal power restored: UPS on line
100403 07/16/98 18:19:40 Self-test at UPS passed
100403 07/30/98 18:19:27 Self-test at UPS passed
Mike Barnett
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Please check out whether you are having any problem with your scsi
cables, Make sure that you are not using more than 5 scsi devices in a
single daisy chain( ie, on one SCSI controller board). Use an ACTIVE
terminator in cases you are using more than 3 SCSI devices on a single
scsi controller.
The scsi devices includes your External HDD,
Internal HDD, Tape Drive, CD-ROM drive etc.
Harish
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Try to turn off the drive and turn it back on again, then run
probe-scsi-all again. If that doesn't work, try to turn it off and
run probe-scsi-all and see if it recognize other disks as usual.
-- Joel Lee
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
try booting of the cdrom and doing an fsck on /dev/rdsk/c1t3d0s0
while booting from cdrom use boot cdrom -sw from the ok prompt. if this
does not fix, try using format command and select the disk you are having
problem with, look at the partition detail or the label info, it might have
been
corrupted, use the backup option to retrive orig lable info.
trying putting this disk offline and boot ( possibly commenting in
/etc/vfstab file)
if other disks work well, then this disk is faulty ( or becoming).
best regards
ganeshan
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Have you made sure the disks are actually spinning up? Possibly
they've blown a fuse or something.
Matt.
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Hi everybody,
I have a SS20 running SunOS 5.4. We had a power outage. When the machine
came back up, I got the message:
/dev/rdsk/c1t3d0s0: No such device or address
Can't open /dev/rdsk/c1t3d0s0
/dev/rdsk/c1t3d0s0: CAN'T CHECK FILE SYSTEM
/dev/rdsk/c1t3d0s0: UNEXPECTED INCONSISTENCY; RUN fsck MANUALLY
Type Ctrl-d to proceed with normal startup,
(or give root password for system maintenance):
I entered the root password and attempted fsck. It went through the 2
interal disks. (the problem disk is 1 of 3 external 4gb artecon ST15230N
Seagate drives)
When it tries to fsck /dev/rdsk/c1t3d0s0, I get:
WARNING: /iommu@f,e0000000/sbus@f,e0001000/esp@0,20000/sd@3,0 (sd18):
logical unit not ready
I brought the machine down to the ok prompt and did a probe-scsi-all and all
the disks show up.
The machine will eventually boot up to multi-user after several logical unit
not ready messages. Once it's up, the external drives do not show up in df.
If I try to mount them manually, I get the warning messages again.
Any help on this would be greatly appreciated,
Loring Safford lsafford@mitretek.org
Mitretek Systems - Z553 phone: (703)610-2818
7525 Colshire Drive fax: (703)610-1603
McLean VA 22102-3481
*********** In a world without fences, who needs Gates? ***********
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:12:46 CDT