README: 2.5 TB Half High LTO-6 Tape Drive Microcode
README: 2.5 TB Half High LTO-6 Tape Drive Microcode
Drive SELF-TEST Procedure
Readme Change History
EU11 - Feature code for internal SAS HHLTO6 drive
8347 - Old feature code for SAS HHLTO6 in 7226 enclosure
8341 - New feature code for SAS HHLTO6 in 7226 enclosure
This package is being provided to allow the update of
2.5TB Half High LTO-6 Tape Drive Microcode
Tape Drives to the latest level of microcode that is currently shipping
from IBM.
Users are responsible to install the latest microcode level.
The microcode and instructions are available
at the IBM external web site:
http://www.ibm.com/support/fixcentral
All levels of AIX that this product was announced on.
The typical time to install this microcode is .2 hrs.
The microcode becomes active when installed and does not require a
system reboot to become active.
Yes, but the tape drive must be available with no tape in the drive,
and no other processing running to the tape drive.
A reboot is not needed as the microcode is active in the drive when the
drive completes microcode download.
In order to complete the microcode download to the tape drive
you will need to use the AIX diagnostics.
The AIX diagnostic package will
need the microcode and dsdata files in this
package.
Note: Atape does not require the dsdata file.
This tape drive package will get you to step where
you will need to use the AIX Diagnostics.
Remove any tape cartridge from the drive prior to attempting the
microcode download.
Do not power off the tape drive during microcode download as this may
damage the tape drive microcode requiring the tape drive to be replaced.
All Power Systems this product was announce on.
In general if the tape drive configures and is available with the VPD
showing the Machine Type and microcode level as described in this
README the machine is one affected.
Note: In some cases you may already have the most current level of
microcode already installed. The following instructions
show you how to check the tape drive microcode level.
To display the level of microcode in the
2.5TB Half High LTO-6 tape drive do the following AIX command:
lscfg -vl rmt0 <--- change the drive number to reflect
the actual rmt address of the 2.5TB
Half High LTO-6 tape drive.
The output of the lscfg command will look like the following with
the:
-
"Device Specific.(Z1)" showing the "Microcode Level":
As of October 2011 there are 2 different P/Ns for the LTO6 drive.
The drives are functionally equivalent and are fully interchangeable.
rmt0 U789C.001.DQDA282-P1-C1-T1-L50000-L0 SAS 1/2-inch
Cartridge Tape Drive (2500000 MB)
Manufacturer................IBM
Machine Type and Model......HH LTO Gen 6
Device Specific.(Z1)........KAJ9
Serial Number...............68000476
Load ID.....................A1700D84
Part Number.................35P0994
FRU Number..................35P1049
EC Level....................M12273A
Device Specific.(Z0)........0180060283011002
Device Specific.(Z3)........N1
The microcode update will bring the IBM 2.5TB Half High LTO6 SAS
tape drive
up to the current level of microcode which is:
KAJ9
Table 1. Microcode History
| Tape Drive P/N
| Tape Drive FRU P/N
| Microcode Level
| Load ID
| Date Released
| What does this microcode change
|
| 35P0994
| 35P1049
| C9T5
| A1700D84
| February 15, 2013
|
-
First Microcode level shipped.
|
| 35P0994
| 35P1049
| D2D3
| A1700D84
| March 22, 2013
|
-
Minimum Level for IBMi Support.
-
Internal RW Channel enhancements
|
| 35P0994
| 35P1049
| D2DB
| A1700D84
| June 25, 2013
|
-
Internal RW Channel enhancements
-
SAS: Performance improvement
-
Fix Cartridge load issue
-
Fix for head position at tape unload
|
| 35P0994
| 35P1049
| D8E5
| A1700D84
| October 4, 2013
|
-
Improve End of Data(EOD) read performance
-
Fix for a read command timeout
|
| 35P0994
| 35P1049
| E4J1
| A1700D84
| May 28, 2014
|
-
Fix problem which caused an FSC 7167 servo perm
-
Servo compensator improvements
-
ERP improvements for FSC 7067 write perms
-
Check if need to invoke a head brush action when entering sleep mode
|
| 35P0994
| 35P1049
| F9A1
| A1700D84
| Dec 2, 2015
|
-
Drive does not come ready after 20 min;
-
Fix of CM read/write problem;
-
Performance degradation (cannot change from speed 14);
-
Panic when entering power down mode (inconsistent power amp);
-
Fix reporting of PEWS;
-
Diag fails with Wrt Protect err if sleep mode;
-
Fix Read Data correction problem;
-
Fix drive head Go To Home Error on HH drives;
-
Fix calibration data corruption after quick insert;
-
Fix No EOD detected issue following extended recovery.
|
| 35P0994
| 35P1049
| G351
| A1700D84
| Jun 12, 2016
|
-
Unexpected LP3 pass count after unload/load
-
Fix MLOI block handling on Writes
-
Fix load hold flush behavior
-
panic due to overlapping flush requests
-
UNLOAD should return GOOD status when cart already ejected
-
fix wrong handling of echo buffer
-
Panic x2E, detection of same dsnum on next wrap
-
LocBot status not suppressed correctly on rewind
-
BFF4 panic due to aborted Write commandMode Unload
|
| 35P0994
| 35P1049
| G9P1
| A1700D84
| December 14, 2016
|
-
Performance improvement
-
Fix queue hang after abort
-
Clear EBC interrupt on drive image start
-
Calibration hang
-
MTR eject doesn't return callback
-
FSC 605E by rewind command after format command fail
-
Fix SAS Hard Reset write state cleanup
-
BOP cache copy should be aborted before flush operation
-
Segment handling for dup dsnum at wrapturn
-
Load (CartInit) fail due to tension shutdown
-
Panic due to extra call to complete function
|
| 35P0994
| 35P1049
| H991
| A1700D84
| September 26, 2017
|
Impact: Availability Severity: ATT
-
Update command timeout values
-
Support DRA using standards based op-codes
-
Support encrypt only (and related) modes
-
Improve drive tape servo handling
-
SCSI command handling and error reporting corrections
-
Log pages 00h and 14h and mode page 10h corrections
-
Improvements in ADT communication
|
| 35P0994
| 35P1049
| J451
| A1700D84
| August 6, 2018
|
Impact: Availability Severity: ATT
-
Improve host interface communications
-
Improve ADI (library interface) communications
-
LN1: not set VS bits in inquiry data
-
Servo control improvements
-
Fixes for log sense and tape alert data
-
Request clean for errors that require re-chuck ERP
|
| 35P0994
| 35P1049
| JAX1
| A1700D84
| December 4, 2018
|
Impact: Availability Severity: ATT
-
Improve log page data reporting
-
Fix medium access queue hangs
-
Correct SCD display for error indications
-
Write Attribute command improvements
-
Error recovery enhancements
|
| 35P0994
| 35P1049
| KAJ9
| A1700D84
| November 4, 2020
|
Impact: Availability Severity: ATT
-
Log Sense data collection and reporting improvements
-
VHF Clean Requested not always set when Tape Alert for Clean is set.
-
Multi-initiator on SAS host interface handling improvements
-
Inquiry page 90h missing second descriptor
-
Mode Sense and Log Sense improvements
-
Reset due to bad link quality
-
Read error recovery handling and enhancements
Write command and write error recovery handling and enhancements
Improved transfer length and overlap handling
Correct parameter checking for Write Attribute command
Host interface logging in and task management message improvements
Allow unload to invoke mid tape recovery after unspool operation failure
Host interface command and abort handling improvements
Error reporting and error recovery improvements
Servo and tape handling enhancements
Improve host interface transfers
Prevent resets that could happen on entering power saving mode or during unload
|
Note: In some cases you may already have the most current level of
microcode already installed. The following instructions
show you how to check the tape drive microcode level.
To display the level of microcode in the
2.5TB Half High LTO6 Fibre Channel
tape drive do the following AIX command:
lscfg -vl rmt0 <--- change the drive number to reflect
the actual rmt address of the 2.5TB
Half High LTO-6 tape drive.
The output of the lscfg command will look like the following with
the:
-
"Device Specific.(FW)" showing the "Microcode Level":
As of October 2011 there are 2 different P/Ns for the LTO6 drive.
The drives are functionally equivalent and are fully interchangeable.
rmt0 U789C.001.DQDA282-P1-C3-T2-W50050763124C4667-L0 IBM
3580 Ultrium Tape Drive (FCP)
Manufacturer................IBM
Machine Type and Model......ULT3580-HH6
Serial Number...............1068000462
Device Specific.(FW)........KAJ9
The microcode update will bring the IBM 2.5TB Half High LTO6
Fibre Channel tape drive
up to the current level of microcode which is:
KAJ9
Table 2. Microcode History
| Tape Drive P/N
| Tape Drive FRU P/N
| Microcode Level
| Load ID
| Date Released
| What does this microcode change
|
| 35P0998
| 35P1060
| C9T5
| None
| February 15, 2013
|
- First Microcode level shipped.
|
| 35P0998
| 35P1060
| D2D3
| None
| March 22, 2013
|
- Internal RW Channel enhancements
- Enhancement for SRR handling
|
| 35P0998
| 35P1060
| D2DB
| None
| June 25, 2013
|
-
Internal RW Channel enhancements
-
Fibre Channel enhancement
-
Fix Cartridge load issue
-
Fix for head position at tape unload
|
| 35P0998
| 35P1060
| D8E5
| None
| October 4, 2013
|
-
Improve End of Data(EOD) read performance
-
Fix for a read command timeout
|
| 35P0998
| 35P1060
| E4J1
| None
| May 28, 2014
|
-
Fix problem which caused an FSC 7167 servo perm
-
Servo compensator improvements
-
ERP improvements for FSC 7067 write perms
-
Check if need to invoke a head brush action when entering sleep mode
|
| 35P0998
| 35P1060
| F9A1
| None
| Dec 2, 2015
|
-
Drive does not come ready after 20 min;
-
Fix of CM read/write problem;
-
Performance degradation (cannot change from speed 14);
-
Panic when entering power down mode (inconsistent power amp);
-
Fix reporting of PEWS;
-
Diag fails with Wrt Protect err if sleep mode;
-
Fix Read Data correction problem;
-
Fix drive head Go To Home Error on HH drives;
-
Fix calibration data corruption after quick insert;
-
Fix No EOD detected issue following extended recovery.
|
| 35P0998
| 35P1060
| G351
| None
| Jun 12, 2016
|
-
Unexpected LP3 pass count after unload/load
-
Fix MLOI block handling on Writes
-
Fix load hold flush behavior
-
panic due to overlapping flush requests
-
UNLOAD should return GOOD status when cart already ejected
-
fix wrong handling of echo buffer
-
Panic x2E, detection of same dsnum on next wrap
-
LocBot status not suppressed correctly on rewind
-
BFF4 panic due to aborted Write commands
-
Fix Ldr Blk park position after Idle Mode Unload
|
| 35P0998
| 35P1060
| G9P1
| None
| December 14, 2016
|
-
Fix queue hang after abort
-
Calibration hang
-
Fix SAS Hard Reset write state cleanup
-
2E0C false detection on thread
-
Segment handling for dup dsnum at wrapturn
-
FSC 78B0 improvement (over rotation at stoplock)
-
Correct FELO/FULO during media read
-
Panic due to extra call to complete function
|
| 35P0998
| 35P1060
| H991
| None
| September 26, 2017
|
Impact: Availability Severity: ATT
-
Update command timeout values
-
Support DRA using standards based op-codes
-
Support encrypt only (and related) modes
-
Improve drive tape servo handling
-
SCSI command handling and error reporting corrections
-
Log pages 00h and 14h and mode page 10h corrections
-
Improvements in ADT communication
|
| 35P0998
| 35P1060
| J451
| None
| August 6, 2018
|
Impact: Availability Severity: ATT
-
Improve host interface communications
-
Improve ADI (library interface) communications
-
LN1: not set VS bits in inquiry data
-
Servo control improvements
-
Fixes for log sense and tape alert data
-
Request clean for errors that require re-chuck ERP
|
| 35P0998
| 35P1060
| JAX1
| A1700D84
| December 4, 2018
|
Impact: Availability Severity: ATT
-
Improve log page data reporting
-
Fix medium access queue hangs
-
Correct SCD display for error indications
-
Write Attribute command improvements
-
Error recovery enhancements
|
| 35P0994
| 35P1049
| KAJ9
| A1700D84
| November 4, 2020
|
Impact: Availability Severity: ATT
-
Log Sense data collection and reporting improvements
-
VHF Clean Requested not always set when Tape Alert for Clean is set.
-
Multi-initiator on SAS host interface handling improvements
-
Inquiry page 90h missing second descriptor
-
Mode Sense and Log Sense improvements
-
Reset due to bad link quality
-
Read error recovery handling and enhancements
Write command and write error recovery handling and enhancements
Improved transfer length and overlap handling
Correct parameter checking for Write Attribute command
Host interface logging in and task management message improvements
Allow unload to invoke mid tape recovery after unspool operation failure
Host interface command and abort handling improvements
Error reporting and error recovery improvements
Servo and tape handling enhancements
Improve host interface transfers
Prevent resets that could happen on entering power saving mode or during unload
|
There are 2 ways a user can get the microcode files.
Each of these will put the microcode on the user's system but the
user may have to move the files to appropriate directory for system
use.
When the user downloads the microcode files from the IBM web site the
packaged microcode files will be put into the
/tmp
directory.
- If the user downloads the
13
HHLTO6.bin
file the user will put the downloaded file into the
/tmp
directory and then use the instructions in the section of this README
titled
"AIX bin file".
- If the user downloads the
TapeHHLTO6-KAJ9-01.rpm
file the user will put the downloaded file into the
/tmp
directory and then use the instructions in the section of this README
titled
"RPM file".
- If the CD-ROM process is used to put the
HHLTO6.bin
file on the user system the microcode files will expanded into the
/tmp/LTO6_web
directory.
The next step is to follow the instructions in this README titled
"AIX bin file"
and copy the files to the appropriate directory.
The building and putting of the files into the
/tmp/LTO6_web
directory was done by the user when they used the CD-ROM build
instructions.
- If the CD-ROM process is used to put the
TapeHHLTO6-KAJ9-01.rpm
file on the user system the microcode files will be in the
tmp/microcode/RPM
directory. To put the microcode in the appropriate directory for the
system the user executes the following AIX commands:
cd /tmp/microcode/RPM
rpm -ihv --ignoreos TapeHHLTO6-KAJ9-01.rpm
(add --force if previous rpm package installed and get conflict error)
This puts the following files in the /etc/microcode directory for
system use:
- HH_LTO_Gen_6.A1700D84.dsdata
- HH_LTO_Gen_6.A1700D84.KAJ9
- HH_LTO_Gen_6.FC.KAJ9
The next step is to follow the instructions in this README titled
"Verifying the Microcode".
Note:
AIX commands and file names are case sensitive.
If the users AIX permission does not allow the copy to the above stated
directory or file then the user will be prompted for a new location.
- Download the
TapeHHLTO6-KAJ9-01.rpm
file to the eServer pSeries to the
/tmp
directory.
- To expand the
TapeHHLTO6-KAJ9-01.rpm
file use the following commands:
cd /tmp
rpm -ihv --ignoreos TapeHHLTO6-KAJ9-01.rpm
(add --force if previous rpm package installed and get conflict error)
- This will put the following files in the
/etc/microcode
directory:
- HH_LTO_Gen_6.A1700D84.dsdata
- HH_LTO_Gen_6.A1700D84.J451
- HH_LTO_Gen_6.FC.KAJ9
To insure the file was downloaded without error do the
following:
- Change to the directory where you just put the microcode.
cd /tmp/LTO6_web
or
cd /etc/microcode
- Use the AIX command sum to verify the checksum of the microcode
file:
sum HH_LTO_Gen_6.*
- Compare the sum you got with the sum of the microcode files here, if
they do not match recheck your steps in creating the file on the
system:
29702 6142 LTO6_web/HH_LTO_Gen_6.A1700D84.KAJ9
17620 2 LTO6_web/HH_LTO_Gen_6.A1700D84.dsdata
04868 6196 LTO6_web/HH_LTO_Gen_6.FC.KAJ9
You need to determine how you are going to download the microcode:
AIX diagnostics
naming and supported directory for the microcode and dsdata file.
- AIX diagnostics requires the microcode file
and the dsdata file to be in the
/etc/microcode
directory.
- For Atape it is not necessary to copy the files to another
directory.
Atape can use any directory. Use the /tmp/LTO6_web directory.
Copy the Dsdata and Microcode files to appropriate directory
for use by AIX diagnostics.
Note: The copy command copies the file to the new directory with the
same name when you use the period(.) after new directory name.
If you downloaded the
RPM
file and executed it the microcode is in
the correct directory with the correct name for AIX diagnostic use.
If you downloaded the .bin
file and executed it then you need to do
the following AIX commands:
cd /tmp/LTO6_web
cp HH_LTO_Gen_6.A1700D84.dsdata /etc/microcode/.
cp HH_LTO_Gen_6.A1700D84.KAJ9 /etc/microcode/.
There are 2 ways to do the microcode download using the AIX
diagnostics.
- Using the AIX Diagnostics menu
- Using the AIX Diagnostics via command line input.
To use the AIX Diagnostics menu for microcode download do the
following:
Note: There are differences between AIX levels of AIX Diagnostics.
In some cases the step after Task selection is Microcode Tools, or
Download Microcode. The following steps are general instructions to
download microcode.
diag (enter)
Task Selection (enter)
Microcode Tasks (enter)
Download Microcode (enter)
Select the drive you wish to download microcode to and follow
the system instructions.
Note: When the system asks where the microcode file is to be
found select
/etc/microcode
The command line to download microcode will use the highest
level of microcode in the /etc/microcode directory for the device
selected. To use the AIX Diagnostics command line input for
microcode download do the following:
Note: If the previous level of microcode is not in the /etc/microcode
directory you will need to use the -f option(force) as the diagnostics
expect the previous level to be present in the /etc/microcode directory.
Note: If the microcode download fails install:
- Insure the checksum is correct for the files that you have downloaded
- If the download to the tape drive fails, use the eject button on the
tape drive to reset the drive. Some microcode download failures leave
the drive in way that the tape drive will not complete a download without
the tape drive being reset.
To reset the tape drive press the eject button and hold the eject button
pressed in for 15 seconds, this will cause a drive reset.
diag -c -d rmt0 -T "download" (enter)
diag -c -d rmt0 -T "download -f" (enter)
Note: Atape must be installed on the system prior to running
AIX diagnostics.
Enter AIX diagnostics and select "Task Selection"
Select "IBM Tape Drive Service Aids"
Select "Microcode load" and fill in the required fields.
Note: The microcode file used by Atape for the Fibre Channel LTO6 tape
drive is:
/tmp/LTO6_web/HH_LTO_Gen_6.FC.KAJ9
The following procedure is designed to allow you to quickly perform a
complete set of diagnostic tests on your Half High LTO-6 tape drive,
without impacting server operation. This 5 < minute test can also be
used to verify good performance of individual LTO6 tape cartridges.
- Verify that a tape cartridge is not loaded in drive.
To unload a
cartridge, press the eject button on the front of the drive.
-
To Start Drive SELF-TEST:
Press and hold the eject button for more than
6 seconds, and less than 20 seconds(at 20 seconds the drive will
do a manual reset and not run the full drive self test),
release the button. The Ready LED
will continue flashing, indicating that the drive is waiting for a
cartridge to be inserted.The drive will wait up to 15 seconds for a
cartridge to be inserted. If no cartridge is inserted the drive will
exit self test.
-
Start the self-test by inserting IBM Test Tape P/N 45E1129 into the
drive. If an IBM Test Tape is not available, you can substitute a
scratch (blank) LTO6 data cartridge.
Note:
Important!
Use a blank cartridge that does
not contain customer data. During this self-test, the cartridge will
be rewritten with a test pattern and any customer data will be destroyed.
Overall test duration is about 5 < minutes.
Note:
-
A cartridge must be loaded within 15 seconds -- otherwise, the drive
will automatically revert back to normal operation. If necessary, return
to step 1 to reenter diagnostic mode.
-
Use a cartridge that is not write-protected. If a write-protected
cartridge is inserted while the drive is in diagnostic mode, the
cartridge will be ejected -- see step below.
-
Self-testing can only be performed using a write-compatible
cartridge type, and with a cartridge that is not
damaged -- see step below.
-
If a cleaning cartridge is inserted while the drive is in diagnostic
mode, cleaning will occur and the drive will then return to normal
operating mode. To reenter diagnostic mode, return to the "Enter
Diagnostic Mode" step.
-
At any time, self-testing can be stopped by pressing the eject button.
After a current test operation is completed, the cartridge will be
ejected and drive will return to normal operation.
-
While self-testing is in progress, the Ready LED will
continue to flash. The following 3 test steps are performed:
-
The initialization
sequence has an approximate duration of 20 seconds.
The Activity LED indicates tape movement.
-
The hardware
test has an approximate duration of 2 minutes. During that
time, a static test is performed on the drive's electrical components,
and proper operation of the cartridge load/unload mechanism is verified.
-
The write/read test
has an approximate duration of 3 minutes. The
Activity LED indicates tape movement.
When self-testing has completed successfully and no problems were
detected,
the cartridge is unloaded from the drive and all LEDs and SCD are off.
Proper function of both the drive and tape cartridge have been verified.
Note: When the Amber Clean LED remains on, it indicates that
self-testing has completed successfully -- but cleaning is required.
Clean the drive by inserting an IBM Cleaning Cartridge (P/N 35L2086).
The drive is no longer in diagnostic mode, and has been returned to
normal operation.
When a drive problem is detected,
the cartridge will remain loaded
inside the drive and the Amber Fault LED ( ! ) will flash.
Following a self-test failure, the tape cartridge will remain loaded
inside drive, and the drive will remain in diagnostic mode. Press the
eject button to unload tape cartridge and return drive to normal
operation.
Repeat self-test using another tape cartridge, If the test passes
replace any defective media.
If the test again failed contact IBM Service for assistance.
When a media problem is detected,
the cartridge will remain loaded
inside the drive, and the Amber Clean LED will flash.
Repeat self-test using another tape cartridge, and replace any defective
media.
When the drive has determined that an incorrect tape cartridge was
inserted, the cartridge is unloaded and both the Activity and
Clean LEDs will flash. Self-testing cannot be performed if
the inserted cartridge is one of the following:
- Write-protected
- Damaged
- Media type that is not write-compatible with the drive
Press the eject button, to end self-test and return the drive to normal
operating mode. Then return to step 1 and run the self-test using a
suitable cartridge.
Following a self-test failure, the tape cartridge will remain loaded
inside drive, and the drive will remain in diagnostic mode. Press the
eject button to unload tape cartridge and return drive to normal
operation.
Table 3. Readme Change History
| Date Changed
| What does this change
|
| May 28, 2014
|
- Update for E4J1 FW release
|
| Oct 9, 2014
|
- Update for adding SAS feature code 8347, 8341
|
| Jul 30, 2015
|
- Update for adding firmware update instructions for IBMi
|
| Dec 2, 2015
|
- Update for F9A1 FW release
|
| Jun 12, 2016
|
- Update for G351 FW release
|
| Dec 14, 2016
|
- Update for G9P1 FW release
|
| Sep 26, 2017
|
- Update for H991 FW release
|
| Nov 28, 2017
|
- Update for H991 FW release
- Add Impact and Severity
|
| Jan 17, 2018
|
- Update firmware update instructions for IBMi
|
| Aug 6, 2018
|
- Update for J451 FW release
|
| Dec 4, 2018
|
- Update for JAX1 FW release
|
| Nov 4, 2020
|
- Update for KAJ9 FW release
|