Only 4 drives are being used instead of 5 drives

[gopi_enovate]

Hi Everyone,
This is my first post in the forum.

Environment Details:

One Master/ Media server and two media servers.

Out of the two media servers, one uses a DELL ML6040 FC tape library. There are 5 FC tape drives present in the library which are not shared with other media servers.
DSSU starts duplication jobs for this media server but only 4 tape drives are being used and one is left free always.

This left tape drive is not the same. It shuffles across different tape drives.

DSSU is done using media manager created using a storage unit group.
Maximum concurrent write drives : 5
Maxmimum stream per drive: 8

MDS allocation is done only for 4 drives at any particular time.

bpstulist for robot storage unit group.
Label: XXXXXXXXX
Storage Unit Type: Media Manager
Host Connection: XXXXXXXXX
Number of Drives: 5
On Demand Only: no
Max MPX/drive: 8
Density: hcart3 - 1/2 Inch Cartridge 3
Robot Type/Number: TLD / 0
Max Fragment Size: 1048576 MB

Staging storage units point to the right media server in "Host connection"
nbrbutil -resetmediaserver <mediaserver> and ltid deamon was reset.

Kindly help me in resolving the issue.

[RamNagalla]

So Do you see any Queue jobs saying that no Drives are avaliable when only 4 Drives are being used and 1 left free?

or no Queue jobs... ?

[gopi_enovate]

Yes, there are queued jobs with status 'awaiting resources, no drives are available'

[Marianne]

Please show us output of nbrbutil -dump? I have seen how an orphaned media allocation caused a similar issue. This old media allocation assumed a drive should be associated with it and always used one drive less than available drives.

[mph999]

I've also seen 'hidden' assigned drives, so they do not show in nbrbutil -dump, but are released with nbrbutil -resetall

So with no jobs running, I would just run nbrbutil -resetall and see if that resolves it.

[gopi_enovate]

Thanks for your reply.
Kindly find MDS allocations in the attached file.

nbrbutil -listorphanedmedia or nbrbutil -listorphaneddrive is not throwing any output.
Believe, there isnt any if it doesnt throw a value.

nbrbutil -resetmediaserver was executed to clear all active resources for the media server.
Is resetall any different than resetting the media server?

In addition to this, ltid daemon was reset on this media server.

[sdo]

Wondering if the tape pool is limited to max 4 partially full media?

Can you share output from this command:

vmpool -list_all

[gopi_enovate]

Kindly find the requested output.
===================================================================
pool number:  0
pool name:    None
description:  the None pool
max partially full:    0
================================================================================
pool number:  1
pool name:    NetBackup
description:  the NetBackup pool
max partially full:    0
================================================================================
pool number:  2
pool name:    DataStore
description:  the DataStore pool
max partially full:    0
================================================================================
pool number:  3
pool name:    CatalogBackup
description:  NetBackup Catalog Backup pool
max partially full:    0
================================================================================
pool number:  4
pool name:    App1
description:  System Files
max partially full:    5
================================================================================
pool number:  5
pool name:    Scratch
description:  Scratch
max partially full:    0
================================================================================
pool number:  6
pool name:    ApplicationFiles
description:  Data Files
max partially full:    5
================================================================================
pool number:  7
pool name:    Exchange-Server
description:  Exchange Files
max partially full:    5
================================================================================
pool number:  8
pool name:    SQL-DB
description:  SQL Databases
max partially full:    5
================================================================================
pool number:  9
pool name:    Exch-DB
description:  Exchange DB
max partially full:    5
================================================================================
pool number:  10
pool name:    ExchangePool
description:  Exch-DB
max partially full:    5
================================================================================
pool number:  11
pool name:    ApplicationSCN
description:  Radar
max partially full:    5
================================================================================
pool number:  12
pool name:    No_Expire
description:  Not Expiring
max partially full:    1
================================================================================
pool number:  13
pool name:    Dedupe
description:  Dedupe
max partially full:    0
================================================================================
pool number:  14
pool name:    ExchangeDB
description:  Exchange-DB
max partially full:    5
================================================================================
pool number:  15
pool name:    ExchangeDB1
description:  Ex Mail Pool for FLO
max partially full:    5
================================================================================
pool number:  16
pool name:    ExchangeDB2
description: 
max partially full:    5
================================================================================
pool number:  17
pool name:    SQL_2012
description: 
max partially full:    5
================================================================================
pool number:  18
pool name:    SP
description: 
max partially full:    5
================================================================================
pool number:  19
pool name:    EVClosed
description: 
max partially full:    5
================================================================================
pool number:  20
pool name:    EV_Pool
description: 
max partially full:    5
================================================================================
pool number:  21
pool name:    Joshna
description: 
max partially full:    7
================================================================================

Please note that these drives were working well before netbackup reboot in this server.
 

[Marianne]

Thanks for nbrbutil output, but without media server name we have no idea which one we are troubleshooting (you chose to blank out media server name in bpstulist output ). nbrbutil -resetall seems like the easiest way out.

[sdo]

What is the output from this on master server:

    bpconfig -L

[gopi_enovate]

eqn-bu3.eqn.local is the media server which has issues.

C:\>bpconfig -L

Mail Admin:            
Job Retry Delay:        10 minutes
Max Jobs/Client:        100
Backup Tries:           2 in 24 hours
Keep Logs:              28 days
Max drives/master:      0
Compress DB Files:      older than 14 days
Media Mnt Timeout:      0 minutes (unlimited)
Shared Timeout:         0 minutes(unlimited)
Display Reports:        1766596718 hours ago
Keep TIR Info:          1 days
Prep Interval:          4 hours (default)
Max Backup Copies:      10
DB Clean Interval:      12 hours
DB Clean Wait Time:     60 minutes
Policy Update Interval: 2 minutes

We will plan a "resetall" as a last restort.

[Marianne]

Thanks. I see only 4 allocations in total for this media server. So, other than possible hidden allocation, I don't see a problem here. Maybe post all the text in a queued job? There may be something there that we have missed.

[gopi_enovate]

Text from a queued job,

5/6/2015 11:25:39 AM - begin Duplicate
5/6/2015 11:25:39 AM - requesting resource eqn-bu3-hcart3-robot-tld-0
5/6/2015 11:25:39 AM - awaiting resource eqn-bu3-hcart3-robot-tld-0 - No drives are available.

Even, i suspect a hidden allocation which we are not able to find out. Will try 'resetall" and let you know if that changes anything.

[gopi_enovate · Accepted solution]

Issue is fixed.
There was a hidden allocation in RB which didn't show up in nrbutil -dump.

However, hidden allocation reported in nbrbutil -reportInconsistentAllocations.

Managed to sync RB and MDS using nbrbutil -syncallocations command.
And Resetmedia server did the trick. I am still not sure why -resetmediaserver first time didnt sync the allocations before. Is it as expected?

Thanks everyone for all your replies.

[Marianne]

Thanks for the feedback!

I all honesty - I have never seen or noticed the 'reportInconsistentAllocations' option.

I have marked your post as solution.