cancel
Showing results for 
Search instead for 
Did you mean: 

how does one control the cleaning cartridge

manatee
Level 6

NBU 7.6.0.3

how does one know if the cleaning cartridges are being used? as i see the "Mounts" column in Media to be always zero, how can check and control this cartridges?

1 ACCEPTED SOLUTION

Accepted Solutions

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified

Your densities do not match.

Tape drives have hcart2 density and cleaning cartridge density is hcart.
Densities in NBU must match.

Change the density of cleaning tapes to 1/2 inch cleaning tape 2 (HC2_CLN)

View solution in original post

22 REPLIES 22

Will_Restore
Level 6

If your tape library suppports self-cleaning (auto clean or whatever they might call it), use it.  The hardware knows better when to clean the drives than NetBackup. 

Review library logs to see when/how often the drives are being cleaned, and number of mounts for the cleaning cartridge.

 

Genericus
Moderator
Moderator
   VIP   

Agree with Will - I have an SL8500 Tape Library which is controlled by an ACSLS - NetBackup sees the tapes, I leave them in the NONE pool. As far as NetBackup is concerned, they have never been mounted. All they have is a creation date when they got added.

I turned off ALL cleaning process within NetBackup, the ACSLS and robot take care of everything automagically.

I did have to set up a script to check the cleaning count on the ACSLS server, to notify the operators to remove and load a new one after 50 uses.

Other than that it works nicely.

 

Perhaps if you told us what type of robot or drives you had, we could give you better advice?

 

 

NetBackup 9.1.0.1 on Solaris 11, writing to Data Domain 9800 7.7.4.0
duplicating via SLP to LTO5 & LTO8 in SL8500 via ACSLS

manatee
Level 6

it's an SL500 (hcart2 drives, type is HP Ultrium) wtih 30 slots. i can see the cleaning cartridges mounted, Active, NONE pool, with the clearnings remaining column at 25. however, the number of mounts is still at zero although the creation column was Oct 2015.

how do you turn OFF cleaning process from NBU?

mph999
Level 6
Employee Accredited

I do NOT recommend turning off cleaning from NBU.  To do this you have to disable TAPE_ALERTS, which disables all TAPE_ALERTS which would mean you may not be alerted to some issue that in the worst case, leads to data loss.

If you run /usr/openv/volmgr/bin/tpclean, this will show if the drives need cleaning, and when they were last cleaned.

If the environment (data center) is very clean, it is quite possible that the drives need cleaning vaer rarely.  The majority of debris in drives, comes from the air.

 

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified
Where is robot control for SL500? ACSLS or TLD? For ACSLS the cleaning should be done by ACSLS. For TLD, my vote is NBU.

manatee
Level 6

robot control? when i go to Devices/Robots, it says "TLD".

i ran that "./tpclean" command and here is the output:

[root@nbu bin]# ./tpclean
Drv    Type      Mount Time  Frequency   Last Cleaned      Comment
***     ****       **********       *********       ****************     *******
  0    hcart2*   1387.5      0                N/A
  1    hcart2*   1441.3      0                N/A

 

seems it's not being used if i'm reading that frequency column right.

mph999
Level 6
Employee Accredited

So, is the robot set to do the cleaning then ?

If not, then it would appear that the drives don't need cleaning (you would see 'needs cleaning' or similar alongside a drive in tpclean output)

Providing the cleaning tapes are labelled correctly, are in the NONE volume pool and have thesame density oas the drives, but with _clean after (eg hcart_clean) then if required, it should work.

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified

Are cleaning tapes in 'normal' slots used for backup tapes or in dedicated maintenance slots in the robot?
'normal' slots means NBU should do the cleaning and media should be configured correctly in NBU (as per Martin's post above).
Dedicated maintenance slots means NBU is unaware of the cleaning tapes and the robot is responsible for cleaning.

If the cleaning tapes are configured in NBU, show us output for one them:

/usr/openv/volmgr/bin/vmquery -m <media-id>

Tape_Archived
Moderator
Moderator
   VIP   

from ./tpclean output I think your cleaning is done by the Tape Library itself.

Go to Tape Library web console/interface and in Drive Operations menu you should find the Tape Drive Clean option, if you want to do the cleaning manually

I have Quantum one and just need to export once Cleaning Tape is expired (50 Mounts) & Import a New cleaning Tape, cleaning is managed by Library.

Check for : Operations => Cleaning Media => Import/Export (This is Quantum Tape Library Web interface & yours should be similar)

By the way, I have HP LTO5 Tape Drives.

manatee
Level 6

@Marianne,

the command gave me this output:

[root@master bin]# ./vmquery -m CLNU04
================================================================================
media ID:              CLNU04
media type:            1/2" cleaning tape (13)
barcode:               CLNU04
media description:     Cleaning Tapes
volume pool:           None (0)
robot type:            TLD - Tape Library DLT (8)
robot number:          0
robot slot:            8
robot control host:    ovmmanager
volume group:          001_00000_TLD
vault name:            ---
vault sent date:       ---
vault return date:     ---
vault slot:            ---
vault session id:      ---
vault container id:    -
created:               Thu 15 Oct 2015 03:31:49 PM AST
assigned:              ---
last mounted:          ---
first mount:           ---
expiration date:       ---
cleanings left:        25
================================================================================

 

and from the activity monitor i noticed a job running from time to time where the job type is "Image Cleanup". does that refer to the tape head cleanup?

 

 

@ Tape_Archvied ,

no i don't notice anything similar to what you have. this is what's bugging me, there's no way to control the cleaning. and that "cleanings left" is always at 25 even a year has passed.

 

mph999
Level 6
Employee Accredited

Image cleanup is not related to cleaning tapes, it is when NBU expires old images.

vmquery for the cleaning tape look correct.

Try this, and I don't usualy recommend this, but just as a one-off lets see ...

Find the drives in the GUI, pick one, right click and select clean now - what happens ???

Also can you attach this file from the media servers (files in on each media server so maybe more than one, if so, please rename)

/usr/openv/netbackup/db/media/errors

The file could have many lines, so please attach, not copy /paste.

This will log any cleaning requests by the drives,so once I convert the tape alerts we should see if the drives have sent any.

Also, on the media server(s), does this file exist :

/usr/openv/volmgr/database/NO_TAPEALERT

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified

Your densities do not match.

Tape drives have hcart2 density and cleaning cartridge density is hcart.
Densities in NBU must match.

Change the density of cleaning tapes to 1/2 inch cleaning tape 2 (HC2_CLN)

mph999
Level 6
Employee Accredited

Opps, missed that, thanks Marianne ...

mph999
Level 6
Employee Accredited

Opps, missed that, thanks Marianne ...

manatee
Level 6

when i tried to select clean now, i get this:

The following error occured while attempting to clean drive
HP.ULTRIUM5-SCSI.000 on master/media server.
daemon failed accepting connection (59).

maybe because a dedup is currently being done?

there is no /usr/openv/volmgr/database/NO_TAPEALERT file exists.

i've attached the contents of the /usr/openv/netbackup/db/media/errors file.

manatee
Level 6

how is it done to change density?

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified
Are you referring to duplication of dedupe backup to tape that is currently running ? Obviously the drive needs to be unoccupied before you can do a manual clean. Have you changed the media density as per my post below? Also - why do you want to manually clean the drive? Have you seen any message saying that the drive needs cleaning? Excessive, unnecessary cleaning can actually wear out a tape drive...

mph999
Level 6
Employee Accredited

This should work

vmchange -new_mt hcart2_clean -m <media ID of cleaning tape>

manatee
Level 6

yes i have changed the media density now.

================================================================================
media ID:              CLNU04
media type:            1/2" cleaning tape 2 (15)
barcode:               CLNU04
media description:     Cleaning Tapes
volume pool:           None (0)
robot type:            TLD - Tape Library DLT (8)
robot number:          0
robot slot:            8
robot control host:    ovmmanager
volume group:          001_00000_TLD
vault name:            ---
vault sent date:       ---
vault return date:     ---
vault slot:            ---
vault session id:      ---
vault container id:    -
created:               Thu 15 Oct 2015 03:31:49 PM AST
assigned:              ---
last mounted:          ---
first mount:           ---
expiration date:       ---
cleanings left:        25

not really want to do manual cleaning but wanting to know why even after a year, the "cleaning left" number is still at 25. shouldn't that number decrement after each usage? i want to find out if the cleaning media is really being used.