Backup drive showing reduction in capacity

[DByrne]

Windows 2003 server sp1 - HP Ultrium storageworks 460.Since last week the backup runs to 72 gb then ejects the tape. I ran HP storageworks & tape tools, it shows that the drive may be showing a reduction in capacity. Now this is happening with all the tapes and a new tape. Previously they have worked fine. I have tried a full erase, but with same results.

[Curtis Ballard]

Post an L&TT support ticket here so we can take a look and see what it is complaining about.

[DByrne]

Here is the support ticket. I uninstalled all the drivers yesterday and rebooted. Same problem. The speed on the backups was always 1400 - 1500 mb/s, now it is around 500 - 400 mb/s.

Support Ticket
Date Generated:
Date Viewed: Thu Dec 14 11:11:35 2006
This Support Ticket contains information about the attached storage devices. It contains essential information, such as device configuration and logs. This information makes solving your problem faster and more reliable. Please generate report and e-mail/fax it to your support provider.
This support ticket was generated at detail level "Normal".



Support Ticket Identifier (b62cdf94-fa9b-4230-96f2-4b52c1ca79f0)
1. System Information
System Model : ProLiant ML370 G4
Operating System : Windows Server 2003 Service Pack 1
Processor : 4 x unknown
System Name : MWL_SRV02
Physical Memory : 2047 MB
Tool Version : Version 4.2
System Time : Thursday December 14 2006 - 11:04:14

1.1 Device Listing: By Connection
Host Bus Adapter 'cpqcissm SCSI Adapter' at Address '2/0.255.0'
Unknown 'COMPAQ SCSI COMMUNICATE' at Address '2/0.0.0'
Disk 'HP LOGICAL VOLUME' at Address '2/0.4.0'
Host Bus Adapter 'cpqcissm SCSI Adapter' at Address '2/1.255.0'
Host Bus Adapter 'cpqcissm SCSI Adapter' at Address '2/2.255.0'
Host Bus Adapter 'cpqcissm SCSI Adapter' at Address '2/3.255.0'
Host Bus Adapter 'cpqcissm SCSI Adapter' at Address '2/4.255.0'
Host Bus Adapter 'cpqcissm SCSI Adapter' at Address '2/5.255.0'
Host Bus Adapter 'Lsi_scsi SCSI Adapter' at Address '3/0.255.0'
Host Bus Adapter 'Lsi_scsi SCSI Adapter' at Address '4/0.255.0'
Tape 'HP Ultrium 2-SCSI' at Address '4/0.4.0'

1.2 Device Listing: By Product
Libraries
Drives
Tape 'HP Ultrium 2-SCSI' at Address '4/0.4.0'
Other Devices
Unknown 'COMPAQ SCSI COMMUNICATE' at Address '2/0.0.0'
Disk 'HP LOGICAL VOLUME' at Address '2/0.4.0'


2. Detailed Device Information
2.1 Tape at address (4/0.4.0)
L&TT Device History
Core Data
Increase the 'Current Detail Level' (at the top of the screen) for further information.
For an explanation of the terms used in the report, please check the on-line L&TT user documentation.
Drive Identity
Drive
Product ID : HP Ultrium 2-SCSI
Format : LTO-2 (w/r), LTO-1 (w/r)
Firmware : F60D/Disty(0x9) (Ver 009.119, Date 2005/10/17-14:17)
Serial number : confidential info erased
SCSI Drive ID : 4
OBDR : Supported
WORM : Not supported

Interface
Drive Capability : SCSI U160
Drive Health

Drive
Device Analysis : Passed with Warnings
version: V06.11.2006
Firmware rev F60D is up-to-date for Ultrium 2-SCSI as of Tue May 23 19:00:00 2006.
Rule 15:
The drive may be showing a reduction in capacity.

Please run the LTO Drive Assessment Test using known good media to verify if the drive is good.
Rule 19:
The drives hardware compression algorithm is currently switched off.
This was probably set by the backup software.
The drive's setup has not been altered.
hp recommends the use of hp storage media.
There were 18 rules checked.
Device Analysis completed, and warnings have been reported.
Write/read (last 3 tapes) : Warning (average 99%, worst <-100%)
Head life : Great margin, 99% life remaining
Reposition life : Great margin, 99% life remaining
Load/unload life : Great margin, 99% life remaining

Drive Configuration
Drive
Data compression : Not enabled
Current block size : 64 KB
Detect WORM tampering : WORM not supported
Reads after WORM tampering : WORM not supported
Media removal (software controlled) : Allowed
Interface
SCSI Drive ID : 4
SCSI Host ID : 7
Negotiated burst data rate : 160 MB/s

Drive Environment
Current temperature : 24.4 C - Great margin (71%)
Maximum temperature since power cycle : 24.7 C - Great margin (73%)
Minimum temperature since power cycle : 23.6 C - Great margin (67%)

Drive Performance
Drive
Estimated native data rate (last 3 tapes) : 12.4 MB/s (104 GB written, 7.5 MB read)
Specified max native data rate : 30 MB/sec.
Max data rate achieved (internal to drive) : Unknown
Average data rate before compression : Unknown
Average data rate after compression : Unknown
Write compression ratio : Unknown
Read compression ratio : Unknown

Drive Usage
Drive
Power on time : 573 days, 8 hours
Tape Pulled : 71.3 full volume equivalents
Load/unload cycles : 271 (0% of expected life)
Duty cycle : 1%
Power cycles : 263
Cleans : 10
Firmware upgrades : 2
Native data volume read/written (last 5 tapes) : Wrote 104 GB, read 7.5 MB

Drive History
Errors returned to host
(no host-reported errors found in the fault logs)

Drive assessment test history
Test last run : 02:02:16:38 power-on (days:hours:minutes:seconds) ago
Firmware revision when last run : F60D
Test options used : Custom
: Device Analysis : Enabled
: SCSI Test : Enabled
: Write / Read Test : Enabled
: Ignore Cleaning Requests : Enabled
Cartridge serial number used : confidential info erased
Final result : Failed
Failed state : 0x12
Predicted margin : Warning (-93%)
Test options (previous test) : Default
: Device Analysis : Enabled
: SCSI Test : Enabled
: Write / Read Test : Enabled
: Ignore Cleaning Requests : Disabled
Final result (previous test) : Passed
Predicted margin (previous test) : Great margin (100%)
Device analysis history

Drive mechanism:
No problems have been identified with the drive mechanism.

Drive PCA:
No problems have been identified with the drive PCA.

Write / Read capability:
Rule Number : 5
Rule Description : Unrecovered read errors
Problem description : The drive heads may be dirty.
Recommended action(s) : The drive has been cleaned (as requested) since the problem was last reported.
: Please try using a different/new cartridge (to check if the problem repeats).
Problem was last reported : 61:02:17:57 power-on (days:hours:minutes:seconds) ago
Rule Number : 15
Rule Description : Capacity loss problems
Problem description : The drive may be showing a reduction in capacity.
Recommended action(s) : Please run the LTO Assessment Test using known good media
: to verify if the drive is good.
Problem was last reported : 02:02:16:38 power-on (days:hours:minutes:seconds) ago
Cartridge problems:
No cartridge related problems were identified with the drive.
Miscellaneous problems:
Rule Number : 19
Rule Description : Performance and configuration check
SCSI configuration (ID0) : Passed
Data Compression configuration (ID1) : Compression Disabled
WWN Check (ID2) : Passed
Cleaning Check (ID3) : Cleaning not requested
Media removal prevented (ID4) : Allowed
Problem was last reported : 02:02:16:38 power-on (days:hours:minutes:seconds) ago
hp recommends the use of hp branded media.
Please run the LTO assessment test to check the operation of your drive.
(loaded tape)
Cartridge Identity
Vendor : HP
Format : LTO-1
Capacity : 100 GB (native), 200 GB (assuming 2:1 compression)
Serial number : confidential info erased
Barcode : Unknown
Manufactured : 07/03/2002
Cartridge Health
Data on tape : Not enough data
Drive measurements : Warning (average 99%, worst <-100%)
Write/read life : Great margin, 83% life remaining (41 full volume equivalents used out of 260)
Load/unload life : Great margin, 99% life remaining (68 load/unloads used out of 20,000)
Cartridge Configuration
Write protected : No
WORM : No
Cartridge Environment
Current temperature in drive : 24.4 C - Great margin (71%)
Maximum temperature in last 3 drives : 25.9 C - Great margin (79%)
Minimum temperature in last 3 drives : 22.6 C - Great margin (62%)
Cartridge Performance
Estimated native data rate (last 1 drives) : 14.5 MB/s (51.5 GB written, 2.8 MB read)
Cartridge Usage
Native capacity used : 29 MB (0%)
Remaining native capacity : 100 GB
Maximum native capacity : 100 GB
Native data volume read/written in last 3 drives : Writes 51.5 GB, reads 2.8 MB
Total native volume : Writes 4.2 TB, reads 87.7 GB
Load/unloads : 68
Cartridge History
Device Analysis
Temperature Measurements
Write Error Rate Log
Read Error Rate Log
Write Error Counters
Read Error Counters
Sequential Access Counters
Tape Usage Log
Tape Capacity Log
Performance Log
Data Compression Log
Fault Log
Event Log
SCSI Command History
SCSI Configuration
ACI Command History
ACI Configuration
ADI Command History
Drive Trace Log
EEPROMS
Mechanical EEPROM
PCA EEPROM
Cartridge Memory

[Curtis Ballard]

I should have asked if you could attach a zipped copy of the support ticket. We can pick a few things out of the text data you posted but I could give you better answers if the whole ticket were available.

[DByrne]

As requested. Today trying different thinks, the drive speed has reduced to 168 mb/s

[Curtis Ballard]

For some reason the L&TT file downloads as a 0 Byte file so I can't tell anything new but there is some data in the original posting. I was hoping to get some more details behind it but there is definitely a problem reported:

Drive measurements : Warning (average 99%, worst <-100%)

That line says that on average the drive has 99% margin which is really good. At 0% margin the drive is operating at the minimum tolerance for acceptable functionality and 100% it is factory new. The worst case margin of <-100% is a problem though. That means that one of the measurements reported a complete failure. I can't tell what measurement reported that without further data but it isn't good no matter where it comes from.

The ticket shows that the drive was cleaned and that there were still errors after cleaning the drive.

It is possible that there could be a tape problem causing this the standard recommendation would be to use a new tape and run the "drive assessment test". The last time that test was run it concluded that there was a problem that could be drive or media and failed. If the test is repeated with a known good or new tape and still fails the drive needs replaced.

You say in the original posting that the problem happens with all tapes and a new tape so it is likely that you have already ruled out the possibility of a media issue.

You could try cleaning the drive several times but with a -100% margin in one reading that probably won't help.

[BR804169]

I have a customer with a similar problem to that reported in this thread. Hopefully you will find a log attached.

In this case the drive is an A7983AT Ultrium 448 internal tape drive. I believe that all cartridges used are HP brand and were new when the device was installed (less than a year ago).

The problem here is simply that this is a worrying development. The customer's data requirements only get to 60/70 GB at the moment and backups appear to be running OK. As for the original poster, this is happening on several (possibly all) of the customer's cartridges.

[Curtis Ballard]

The drive log just posted by Rod Hewitt shows that one channel in the drive head may have failed. Clean the drive a few times and then use L&TT to run the assessment test.

[fanlabiz78]

Addendum to previous feedbacks:

I'm an old HP LTO user (LTO1-2-3-4), currently using LTO5 (Ultrium 3000).

I was also experiencing backup issues with tape capacity reduction by 20, 30 or even 50%. 

I tried multiple tape cleans - no way. My backups never reached anymore the usual 750 GiB minimum.

When testing with HPE T&TT, I typically found the "drive measurements: warning" in the test log.

I finally solved my issue by discarding the C7978A cleaning tapes I was using for my old LTO units.

These orange cleaning tapes show "all ultrium LTO1, LTO2, LTO3, LTO4" - not LTO5. My bad.

A single cleaning with the appropriate cleaning tape, and my Ultrium 3000 now works like before (like a charm).