Category: Hardware

AIX – Some tape commands

Getting information about the tape drive.

root@aix:/ # lscfg -vpl rmt0
rmt0             U787E.001.AAA2714-P2-C1-T2-L0-L0  4.0 GB 4mm Tape Drive
Manufacturer…………….ARCHIVE
Machine Type and Model……IBM4326NP/RP  !D
Device Specific.(Z1)……..5ALG
Serial Number……………DR0N6QG 0011
Load ID…………………0011
Part Number……………..21H5154
FRU Number………………59H3481
EC Level………………..E30312
Device Specific.(Z0)……..0180020283000018
Device Specific.(Z3)……..L1
PLATFORM SPECIFIC
Name:  st
Device Type:  byte

Getting information about the tape. It is using compression and a variable length block size.

root@aix:/ # tctl -f /dev/rmt0 status
rmt0 Available 06-08-01-0,0 4.0 GB 4mm Tape Drive
ttribute  value description                          user_settable
block_size 0     BLOCK size (0=variable length)       True
compress   yes   Use data COMPRESSION                 True
mode       yes   Use DEVICE BUFFERS during writes     True
ret_error  no    RETURN error on tape change or reset True

Rewinding tape drive

root@aix:/ # tctl -f /dev/rmt0 rewind

Reading tape and redirecting output to a file. In this case, the operating system showed that the tape itself is damaged.

root@aix:/ # tctl -f /dev/rmt0 read > /tmp/testtape
tctl: 0511-578 Cannot read 512 bytes: The media surface is damaged.

Fujitsu hardware running Solaris

Fujitsu made some SPARC machines sold under the brand PRIMEPOWER. It runs Solaris but to see the hardware information you need to use some specific tools that Fujitsu developed.

How to list hardware status

root@solaris8-fujistu:/ # /opt/FJSVmadm/sbin/hrdconf -l

Machine Administration tool is an interactive tool that can also be used to display hardware status

root@solaris8-fujistu:/ # /opt/FJSVmadm/sbin/madmin

If you need to run the equivalent of the Sun Explorer, you’re looking for fjsnap. First, check if you have it installed.

root@solaris8-fujistu:/ # pkginfo -l FJSVsnap
PKGINST:  FJSVsnap
NAME:  System Information Output Tool
CATEGORY:  application
ARCH:  all
VERSION:  2.5,REV=2005.01.1100
BASEDIR:  /
VENDOR:  FUJITSU/PFU LIMITED
DESC:  System Extensive Data & Information Output Tool
PSTAMP:  flicker120050419161312
INSTDATE:  Nov 27 2006 18:22
STATUS:  completely installed
FILES:      182 installed pathnames
3 shared pathnames
14 directories
147 executables
1437 blocks used (approx)

Then run it with /opt/FJSVsnap/bin/fjsnap -a <filename>

root@solaris8-fujistu:/ # /opt/FJSVsnap/bin/fjsnap -a /tmp/`hostname`.tar
<<< fjsnap * START !! >>>
<<< Path-name check * START !! >>>
<<< Path-name check *  END  !! >>>
<<< File-size check * START !! >>>
<<< File-size check *  END  !! >>>
<<<  needed tmp-space size  >>>
12806 MB is left in /var/tmp
143 MB is needed
<<< Copy to tmp-dir * START !! >>>
<<< Copy to tmp-dir *  END  !! >>>
<<< Output * START !! >>>
/usr/sbin/tar cfb /tmp/solaris8-fujistu.tar
<<< Output *  END  !! >>>
<<< fjsnap *  END  !! >>>

To get the machine model you can check

root@solaris8-fujistu:/ # /opt/FJSVhwr/sbin/prtdiag
System Configuration:  Fujitsu  sun4us Fujitsu PRIMEPOWER650 1-slot 4x SPARC64 V

root@solaris8-fujistu:/ # /opt/FJSVhwr/sbin/getmodelcode
650

How to get Primepower server serial number from the system console:
For PRIMEPOWER 250, PRIMEPOWER 450, PRIMEPOWER 650 and PRIMEPOWER 850

root@solaris8-fujitsu:/ # /opt/FJSVmadm/sbin/serialid -a
serialid: 6J123456,123456

For PRIMEPOWER 900, PRIMEPOWER 1500 and PRIMEPOWER 2500

root@solaris8-fujitsu:/ # /opt/FJSVcsl/bin/serialid -m machine_name -a
serialid: 6J123456,123456

The serial number starts with 6J or 6B

Discovering new disks in a Solaris server with QLogic HBA

This server is using the QLogic’s FC driver and not the usual Sun’s driver.

root@solaris9:~ # modinfo | grep ql
25  12beb8f  83803 281   1  qla2300 (QLogic FC Driver v4.18)
307 7876c000  17e6f 280   1  qlog (VxQLOG 4.1_REV-4.1B18_sol_GA_s1)

To discover the new LUNs attached, use the qlreconfig utility

root@solaris9:~ # /opt/QLogic_Corporation/drvutil/qla2300/qlreconfig -d qla2300 -v

*****                     *****
*****    W A R N I N G    *****
*****                     *****

PLEASE READ THE DOCUMENTATION BEFORE USING THIS UTILITY

This utility updates the driver configuration parameters
on an live system.  Extreme care MUST be taken to insure
that the specified parameters are correct for this system.
Failure to do so may render this system unusable, and/or
panic the system.

This utility utilizes the devfsadm utility, which depending on
the system and the number of devices, may take some time to
complete.

Do you wish to continue (y/n)? y
Dec 16 14:50:47  /opt/QLogic_Corporation/drvutil/qla2300/qlreconfig: starting execution
Dec 16 14:50:47  /opt/QLogic_Corporation/drvutil/qla2300/qlreconfig: utility verson is: 1.01
Dec 16 14:50:47  /opt/QLogic_Corporation/drvutil/qla2300/qlreconfig: driver version: 4.18
Dec 16 14:50:47  /opt/QLogic_Corporation/drvutil/qla2300/qlreconfig: os level is: 9
Dec 16 14:50:47  /opt/QLogic_Corporation/drvutil/qla2300/qlreconfig: system call checks done
Dec 16 14:50:47  /opt/QLogic_Corporation/drvutil/qla2300/qlreconfig: Found qla2300 driver ioctl device path at: ‘/devices/pci@1e,600000/SUNW,qlc@2:devctl’
Dec 16 14:50:47  /opt/QLogic_Corporation/drvutil/qla2300/qlreconfig: buffer allocated and populated
Dec 16 14:50:47  /opt/QLogic_Corporation/drvutil/qla2300/qlreconfig: finished with /usr/bin/pkill call
Dec 16 14:50:47  /opt/QLogic_Corporation/drvutil/qla2300/qlreconfig: calling driver ioctl
Dec 16 14:50:48  /opt/QLogic_Corporation/drvutil/qla2300/qlreconfig: ioctl OK — status is: 0; detail status is 0
Dec 16 14:50:48  /opt/QLogic_Corporation/drvutil/qla2300/qlreconfig: finished with driver call
Dec 16 14:50:48  /opt/QLogic_Corporation/drvutil/qla2300/qlreconfig: starting /usr/sbin/devfsadm call
Dec 16 14:51:15  /opt/QLogic_Corporation/drvutil/qla2300/qlreconfig: finished with /usr/sbin/devfsadm call
Dec 16 14:51:15  /opt/QLogic_Corporation/drvutil/qla2300/qlreconfig: starting /usr/sbin/update_drv call
Dec 16 14:51:25  /opt/QLogic_Corporation/drvutil/qla2300/qlreconfig: finished with /usr/sbin/update_drv call
Dec 16 14:51:25  /opt/QLogic_Corporation/drvutil/qla2300/qlreconfig: finished

Resetting a RSC / ALOM Card from Solaris

Here an RSC card wasn’t responding so my colleague decided to reboot the card.

root@solaris9:/ # /usr/platform/`uname -i`/rsc/rscadm shownetwork
rscadm: RSC firmware not responding~

When you pass the flag -s you try a soft reset in the card.

root@solaris9:/ # /usr/platform/`uname -i`/rsc/rscadm resetrsc -s
Are you sure you want to reboot RSC (y/n)?  y
rscadm: RSC firmware not responding

Here is the how to perform the hard reset

root@solaris9:/ # /usr/platform/`uname -i`/rsc/rscadm resetrsc
Are you sure you want to reboot RSC (y/n)?  y

If you need to reset the ALOM card, use scadm with the same parameters as above

root@solaris9:/ # /usr/platform/`uname -i`/sbin/scadm resetrsc -s

Unfortunately, resetting the card didn’t solve the problem.

How to perform a Sun SC Failover

Check if there is any data synchronization running between the System Controllers. The File Propagation State must be set as active and there must not show any file in Active File or Queued Files.

e20k-sc0:sms-svc> showdatasync
File Propagation State: ACTIVE
Active File:            –
Queued Files:           0

Each system controller must not have any status different from Good to prevent the failover. The failover must be active.

e20k-sc0:sms-svc> showfailover -v
SC Failover Status:     ACTIVE
Status of Shared Memory:
HASRAM (CSB at CS0):     …………………………………Good
HASRAM (CSB at CS1):     …………………………………Good

Status of e20k-sc0:
Role:                    …………………………………MAIN
SMS Daemons:             …………………………………Good
System Clock:            …………………………………Good
Private I2 Network:      …………………………………Good
Private HASRAM Network:  …………………………………Good
Public Network:
Group “C1”:      …………………………………..Up
eri0:              …………………………………..Up
eri3:              …………………………………..Up
Logical IP Addr. – C1:…………………………………..Up
System Memory:           ………………………………….4.9%
Disk Status:
/:                   ………………………………….3.6%
Console Bus Status:
EXB at EX0:          …………………………………Good
EXB at EX1:          …………………………………Good
EXB at EX2:          …………………………………Good
EXB at EX3:          …………………………………Good
EXB at EX4:          …………………………………Good
EXB at EX5:          …………………………………Good
EXB at EX6:          …………………………………Good
EXB at EX8:          …………………………………Good
EXB at EX9:          …………………………………Good
EXB at EX10:         …………………………………Good
EXB at EX11:         …………………………………Good
EXB at EX12:         …………………………………Good
EXB at EX13:         …………………………………Good
EXB at EX14:         …………………………………Good
EXB at EX15:         …………………………………Good
EXB at EX16:         …………………………………Good
EXB at EX17:         …………………………………Good

Status of e20k-sc1:
Role:                    ………………………………..SPARE
SMS Daemons:             …………………………………Good
System Clock:            …………………………………Good
Private I2 Network:      …………………………………Good
Private HASRAM Network:  …………………………………Good
Public Network:
Group “C1”:      …………………………………..Up
eri0:              …………………………………..Up
eri3:              …………………………………..Up
Logical IP Addr. – C1:……………………………..Inactive
System Memory:           ………………………………….4.6%
Disk Status:
/:                   ………………………………….3.5%
Console Bus Status:
EXB at EX0:          …………………………………Good
EXB at EX1:          …………………………………Good
EXB at EX2:          …………………………………Good
EXB at EX3:          …………………………………Good
EXB at EX4:          …………………………………Good
EXB at EX5:          …………………………………Good
EXB at EX6:          …………………………………Good
EXB at EX8:          …………………………………Good
EXB at EX9:          …………………………………Good
EXB at EX10:         …………………………………Good
EXB at EX11:         …………………………………Good
EXB at EX12:         …………………………………Good
EXB at EX13:         …………………………………Good
EXB at EX14:         …………………………………Good
EXB at EX15:         …………………………………Good
EXB at EX16:         …………………………………Good
EXB at EX17:         …………………………………Good

You run setfailover force to perform the failover. If there is a problem on the spare clock input, it can cause a domain stop (it is going to power off all the domains)

e20k-sc0:sms-svc> setfailover force
Forcing failover. Do you want to continue (yes/no)? yes
The spare clock input on some boards might be bad. Forcing a failover now is likely to cause the affected domains to domain stop (Dstop).
Do you want to continue (yes/no)? yes

Checking service status

e20k-sc0:sms-svc> showfailover -r
SPARE

When you manually failover, the automatic failover is disabled

e20k-sc0:sms-svc> showfailover –v | grep “SC Failover Status”
SC Failover Status:     DISABLED

You need to manually activate the automatic failover again to return to the previous status

e20k-sc0:sms-svc> setfailover on
e20k-sc0:sms-svc> showfailover -v | grep “SC Failover Status”
SC Failover Status:     ACTIVATING

e20k-sc1:sms-svc> showfailover -v | grep “SC Failover Status”
SC Failover Status:     ACTIVE

SteelSeries Siberia USB Soundcard on Windows 7

I bought a SteelSeries 5Hv2 USB on ProGaming Computer. It’s a nice headset and comes with a USB soundcard similar to the SteelSeries Siberia USB soundcard.

According to the website, it had drivers for Windows XP and Windows Vista but no  support for Windows 7. I solved this problem by installing a beta driver available at the SteelSeries blog. Other people reported success right-clicking the driver installer and selecting to execute in Windows Vista combatibility mode.

Now you just need to download the official Windows 7 drivers at the manufacturer’s website.

Sun System Management Services – Error: Exclusive session is in use, disconnecting.

You need to connect to a System Controller (SC) to manage a domain in a Sun Fire E12K, E15K, E20K or E25K. It has the System Management Services (SMS) software that allows you to control the domains.

To view the console, you use /opt/SUNWSMS/bin/console -d <domain letter or domain name> but I was having the error message “Exclusive session is in use, disconnecting”.

sms-svc@sc0:/ $ console -d domain06
Trying to connect…
Connected to Domain Server.

Exclusive session is in use, disconnecting.

I asked to a friend why I was having this problem and he said that I needed to kill the other session.

sms-svc@sc0:/ $ w
9:24am  up 76 day(s),  3:53,  2 users,  load average: 0.91, 2.04, 1.52
User     tty           login@  idle   JCPU   PCPU  what
root     pts/1         9:08am     9                console -d domain06
root     pts/2         9:19am                      w

I found out which PID is running the console, then killed it.

sms-svc@sc0:/ $ ps -ef | grep consol
root  3127     1  0   Sep 19 console  0:00 /usr/lib/saf/ttymon -g -h -p sc0-01-vix console login:  -T sun -d /dev/console
sms-svc 11229 11182  0 09:24:18 pts/2    0:00 grep consol
sms-svc  5492 5398  0 09:09:23 pts/1    0:00 console -d domain06

sms-svc@sc0:/ $ kill 5492

Worked as expected:

sms-svc@sc0:/ $ /opt/SUNWSMS/bin/console -d domain06
Trying to connect…
Connected to Domain Server.
Your console is in exclusive mode now.

Cray Upgrades “Kraken” Supercomputer with Six-Core AMD Opteron Processors

The third Supercomputer according to the Top 500 List is getting a CPU upgrade by Cray. The AMD Quad-Core processors are being replaced by Six-Core processors without having to replace any board.

Check OBP variables without rebooting Solaris or going to the ok prompt

To check the OBP variables in your Solaris SPARC server and leaving Solaris up and running, you’ll need to use the eeprom command

root@solaris:/ # eeprom
asr-policy=normal
test-args: data not available.
diag-passes=1
local-mac-address?=true
fcode-debug?=false
scsi-initiator-id=7
oem-logo: data not available.
oem-logo?=false
oem-banner: data not available.
oem-banner?=false
ansi-terminal?=true
screen-#columns=80
screen-#rows=34
ttyb-rts-dtr-off=false
ttyb-ignore-cd=true
ttya-rts-dtr-off=false
ttya-ignore-cd=true
ttyb-mode=9600,8,n,1,-
ttya-mode=9600,8,n,1,-
output-device=ttya
input-device=ttya
auto-boot-on-error?=true
error-reset-recovery=sync
load-base=16384
auto-boot?=true
network-boot-arguments: data not available.
boot-command=boot
diag-file: data not available.
diag-device=disk0:a
boot-file: data not available.
boot-device=vx-rootdisk vx-rootmirr
use-nvramrc?=false
nvramrc=devalias vx-rootdisk /pci@1c,600000/scsi@2/disk@0,0:a
devalias vx-rootmirr /pci@1c,600000/scsi@2/disk@3,0:a
security-mode=none
security-password: data not available.
security-#badlogins=0
verbosity=normal
diag-trigger=error-reset power-on-reset
service-mode?=false
diag-script=normal
diag-level=max
diag-switch?=false

HP-UX – fcmsutil – Error: FCP driver not loaded

I saw a message in syslog to check for an EMS Event Notification.

root@hp-ux:/ # cat /var/adm/syslog/syslog.log
Jul 23 11:33:07 hp-ux EMS [2596]: —— EMS Event Notification ——   Value: “SERIOUS (4)” for Resource: “/adapters/events/TL_adapter/0_2_1_0″     (Threshold:  >= ” 3″)    Execute the following command to obtain event details:   /opt/resmon/bin/resdata -R 170131458 -r /adapters/events/TL_adapter/0_2_1_0 -n 170131457 -a

root@hp-ux:/ # /opt/resmon/bin/resdata -R 170131458 -r /adapters/events/TL_adapter/0_2_1_0 -n 170131457 -a

CURRENT MONITOR DATA:

Event Time……….: Thu Jul 23 11:33:07 2009
Severity…………: SERIOUS
Monitor………….: dm_TL_adapter
Event #………….: 23
System…………..: hp-ux

Summary:
Adapter at hardware path 0/2/1/0 : Fibre Channel driver has gone into
non-participating mode
Description of Error:
lbolt value: 3207

The Fibre Channel Driver has gone into non-participating mode
Frame Manager Status Register =  0x200002ff
Probable Cause / Recommended Action:

A cable or hardware problem exists. This can also be caused due
to the Tachyon TL adapter changing its AL_PA across a LIP.
Do the following:
1. Verify that the fiber optic cable is functional and securely
attached to the Fibre Channel Mass Storage Adapter.
2. After checking the cabling, execute a loopback test using
fcmsutil.
Command syntax is as follows:
/opt/fcms/bin/fcmsutil <device_file> lb plm
If the problem persists after verifying fiber optic cabling,
replace the adapter

Additional Event Data:
System IP Address…: 10.122.100.24
Event Id…………: 0x4a6874a300000000
Monitor Version…..: B.01.00
Event Class………: I/O
Client Configuration File………..:
/var/stm/config/tools/monitor/default_dm_TL_adapter.clcfg
Client Configuration File Version…: A.01.00
Qualification criteria met.
Number of events..: 1
Associated OS error log entry id(s):
0x4a68739100000000
Additional System Data:
System Model Number………….: 9000/800/rp4440
OS Version………………….: B.11.11
EMS Version…………………: A.04.20
STM Version…………………: A.49.00
Latest information on this event:
http://docs.hp.com/hpux/content/hardware/ems/dm_TL_adapter.htm#23

v-v-v-v-v-v-v-v-v-v-v-v-v    D  E  T  A  I  L  S    v-v-v-v-v-v-v-v-v-v-v-v-v

Component Data:
Physical Device Path….: 0/2/1/0
Vendor Id……………: 0x0000103C
Serial Number(WWN)……: 50060B00005FCE46

I/O Log Event Data:

Driver Status Code………………: 0x00000017
Length of Logged Hardware Status….: 0 bytes.
Offset to Logged Manager Information: 0 bytes.
Length of Logged Manager Information: 61 bytes.

Manager-Specific Information:

Raw data from FCMS Adapter driver:
00000003 00000C87 00000001 00000001 200002FF 2F75782F 6B65726E 2F6B6973
752F544C 2F737263 2F636F6D 6D6F6E2F 7773696F 2F74645F 6973722E 63

Checking the device file.

root@hp-ux:/ # ioscan -fnC fc
Class     I  H/W Path  Driver S/W State   H/W Type     Description
=================================================================
fc        0  0/2/1/0   td   CLAIMED     INTERFACE    HP Tachyon XL2 Fibre Channel Mass Storage Adapter
/dev/td0
fc        1  0/6/1/0   td   CLAIMED     INTERFACE    HP Tachyon XL2 Fibre Channel Mass Storage Adapter
/dev/td1

I tried to pass td1 as argument but it needed the full path.

root@hp-ux:/ # /opt/fcms/bin/fcmsutil td1 lb plm
Error: FCP driver not loaded.

I passed the entire path and the fcmsutil worked:

root@hp-ux:/ # /opt/fcms/bin/fcmsutil /dev/td0 lb plm

WARNING: This is a DESTRUCTIVE test.
This might result in failure of current I/O requests.
Do you want to continue? (y/n) [n]…