Netbackup Accelerator Error

Try with out "allow multiple datastreams" selected. so there will be only 1 job. Also, what the deduplication setting for the client, media server or client-side? Client-side sometimes causes status 13.
 

This looks like Windows 'Change Journal' issue and not really Acceleraror.

Please see if this section: “Accelerator and the Windows change journal” on page 694 of NetBackup Administrator's Guide, Volume I  helps.
 

Hello Marianne,

Sorry for checking late, as i was out of offive for few days.

I tried backup without accelerator, though backup speed was fast, still received error with code 24/13 and backup failed.

Now a days, i am experienceing backkup failure for all file servers with drive size around TBs failing woth code 13 and 24.

Not able to figure out cause of issue as other file servers with less drives size is working fine.

On both the servers, i have installed Netbackup 7.6.0.1 client version and storage target is DataDomain.

Multistreaming is allowed.

Backup runs fine for some time a then after wrting around 1TB or more it fails it incomplete with code 13/24.

Kindly assist me.

It would seem to me that your file servers have a high rate of change.  This is borne out by the NTFS Change Journal running out of room to track all file changes:

Error for code 13 :  not using change journal data for <F:\>:
unable to initialize change journal usage
<reason=record for requested start usn has been dropped
from the change database (too many changes)>

Can I ask whether the NTFS volumes are quite full too?

If so, then there's probably not a lot you can do from NetBackup side.  Before doing anythign else, I would suggest monitoring both free disk space and VSS disk space consumption and which volumes (drive letters) are being used by which other volumes (drive letters) as VSS delta snap space... on the nodes with failing backup... do this during the actual backups, because this may give you the evidence that you need to present to management to prove that you may need to look at providing some additional storage space, not for files, and not for allplcations - but purely sitting there unused (most of the time) to provide nothing other than delta snap space for VSS.  i.e. you may need to give more space to VSS and to manually configure VSS so as not to leave VSS to make its own mind up about what it should do.

Remember:  NetBackup uses bpfis to call VSS and if VSS fails, then it is most definitely not a NetBackup software issue per-se - in fact, it is a OS and/or server administration and/or storage administration issue.  If VSS cannot do its job, then NetBackup Client cannot do its job.

.

Have you considered FlashBackup instead?  FlashBackup is a backup method which is sometimes a useful alternative (to walking the file system) for volumes which are large and quite full and contain millions of folders and files. 

Hi Sdo

While using accelerator i also received same error " 

<reason=record for requested start usn has been dropped
from the change database (too many changes)>

i disabled "use change journal option" and tried backing up without accelerator, still seeing failure with code 13/42.

I could not understand troubleshooting steps given by you. kindly simplify.

Also, please show us output from the problem client:

wmic logicaldisk list brief

vssadmin list providers

vssadmin list volumes

vssadmin list writers

vssadmin list shadowstorage

vssadmin list shadows

sure ..i will provide the output

one more behaviour , the clients are configured as cluster say client1A and client1B and are windows2012 servers

have configured both the clients in backup.

while backing up one client , its total size for one drive is 5 TB used space is 2 TB. however my backup stream for that drive is running more than3.5 TB dont know how..

have anyone came across this issue.

1) Is NTFS native de-duplication being used within the NTFS volumes?

see:

http://blogs.technet.com/b/filecab/archive/2012/05/21/introduction-to-data-deduplication-in-windows-server-2012.aspx

.

2) Is Enterprise Vault configured to archive the contents of the file system?  Maybe something is missing from your server configuration which is causing stubbed/offline files to be recalled?

.

3) Or maybe you have a different archiving solution, which is causing files to be fully recalled/retrieved from the archive store, i.e. the 'offline-ness' or 'stubby-ness' of the stubbed files is being ignored when NetBackup walks down through the file system looking for files to backup.

e.g.:

https://www-secure.symantec.com/connect/forums/ignoring-stub-files

we have DataDomain system as target device.

We dont have archviing solution , its a standard file system backup.