Problem with fsdmn

[Enkli]

Hello we have a DS20E Tru64 V.5.1 that today mornging we found drives with error leds and the file domains unmounted
Restarted the server but nothing , these were the disk of mirror. But now they ar not blinking red and FSDMN not showing as they say

showfsets: can't show fileset info for domain 'database_dmn'
showfsets: error = I/O error

Any idea?

Replacing any of the hard drives could solve the problem.
The mirror consists of two HDD
Thank you

[Rob Leadbeater]

Hi,

You've given us very little to go on...

How are/were the drives mirrored ?
Any RAID controllers, if so what type ?

The output of:

# sizer -v
# hwmgr view dev
# ls -l /etc/fdmns/database_dmn

would be useful to see...

From the SRM prompt (P00>>>) the output of the commands

show dev
show config

would be useful too.

Regards,
Rob

[Enkli]

Excuse me but I am not now at my workplace. I should have done as you mentioned above.
The system is 5.1B-4
the link inside database_dmn is "dsk2a"

but I cannot post anything else from the other commands.
I can do that tomorrow morning when I get back to work.
Tomorrow I am thinking placing only one new drive on the server and retry?
Or you have any other idea?

I dont know anything about RAID on this system, but as I see it is not done through LSM

The system has 6 HDD 36GB divided in three mirrors , two in each one. so the Disk2 has error. the other disks are reporting ok.

[Rob Leadbeater]

Hi,

OK, post the information when you can...

Personally I wouldn't touch any hardware until you've determined exactly what has broken.

There are a number of different RAID cards that go in a DS20E...

Regards,

Rob

[Enkli]

HWmgr view dev:

4: /dev/dmapi/dmapi
5: /dev/scp_scsi
6: /dev/kevm
46: /dev/disk/floppy0c 3.5in floppy fdi0-unit-0
63: /dev/disk/dsk0c i2o_bs iop-0-dev-0
64: /dev/disk/dsk1c i2o_bs iop-0-dev-1
65: /dev/disk/dsk2c i2o_bs iop-0-dev-2
66: /dev/disk/cdrom0c TEAC CD-224E bus-1-targ-0-lun-0
67: /dev/random
68: /dev/urandom
69: /dev/ntape/tape0 COMPAQ SDT-10000 bus-0-targ-6-lun-0
When I will be at the console I can post the results of other commands

[Enkli]

Hello,
As you suggest I executed the comand "show dev" and the output is:
dqa 0.0.0.105.0 DQA0 CD-224E 9.9A
dva0.0.0.0.0 DVA0

dza524.0.0.2004.0 DZA524(BSA)

dza525.0.0.2004.0 DZA525(BSA)

dza526.0.0.2004.0 DZA526(BSA)

mka600.6.0.7.1 MKA600 compaq SDT-10000

eia0.0.0.3004 EIA0 00-0B-CD-BA-F4-54

eia0.0.0.3005 EIB0 00-0B-CD-BA-F4-55

PKA0.7.0.7.1 PKA0 SCSI BUSS ID 7
PKB0.7.0.6.1 PKC0 SCSI BUSS ID 7
PKC0.7.0.106.1 PKC0 SCSI BUSS ID 7
PZA0.0.0.0.2004 IZ0A0

The output of "Show config" command is:

SRM console V6.4-11
Processors CPU0, CPU1 both Alpha EV68H 2.2 833 MGH
Memory 1 GB

Bus 00 slot 06/0 Adaptec AIC-7895 pkb0.7.0.6.0 SCI Bus ID 7

Bus 00 slot 06/1 Adaptec AIC-7895 pkc0.7.0.106.0 SCI Bus ID 7

Bus 00 slot 07: NCR 53C895
pka0.7.0.7.0 SCI Bus ID 7
mka 600.6.0.7.1 Compaq STD-10000

Thank you in advance,
Enkli

[Kapil Jha]

From your output it seems that server can not see disk other than root disk,
have you tried to boot the server in single user mode
>>> b -fl s

BR,
Kapil+

[Kapil Jha]

you said u have 6 disks but hwmgr command is showing only 2 disks.

BR,
Kapil+

[John Welsh_2]

Hi, From the above hwmgr view dev, the I2o
device is a KZPCC Raid controller.
So 6 physical drives - 3 mirrors - 3 logical
disks.
You should be able to boot into the KZPCC
configurator from the >>> prompt, find out which drive has failed and then replace it.
Regards, John Welsh.

[Enkli]

yes the idea might be this.

but 3 disks seem to fail.
2 of them are forming one logical drive and with that i might be that data is lost.
I am trying to recover but i think there might be any corruption on the controler, because there is two low possibility to have three disks failed at once.
I think i will remove the controller the batery jus for 5 min. and the put it again also a firmware upgrade

if any other idea please inform me
RG