cancel
Showing results for 
Search instead for 
Did you mean: 

Restore takes too long

starra
Level 2

Our restore stuck in Begin reading. is it normal?

5 REPLIES 5

Tousif
Level 6

Hello,

Some time it not show other detail in Job Detail status.

If job detail window showing progress in KB written and files it mean the restore is in progress.

You can also check the bpbrm, bptm and tar log for more information.

 

Thanks 

Marianne
Level 6
Partner    VIP    Accredited Certified

'Begin reading' means the bptm process has started reading the 'tar-file' on storage. When the particular data is located in the tar-file, the data will be read and sent to the relevant process on the client (tar for file-level restore). 

We need all relevant info in order to tell if this is normal or not.

What do you define as being 'stuck'? 
How long since the 'begin reading' message?
Are Kbytes in the job growing?

What type of restore? 
Filesystem or database?
Large amount of data to be restored or small amount from a large backup?
What is the fragment size in the storage unit? 

Please always start a new discussion for your unique issue (not tag onto a 2-year-old discussion that was solved) and give us as much information as possible.

 

Genericus
Moderator
Moderator
   VIP   

How long has it been stuck?

There should be timeouts set that will either start the restore or kill it after a certain time.

There is a feature in NetBackup that MAY prevent your restore if these values are not set right in the master bp.conf

what are the values if any for 

MPX_RESTORE_DELAY =

 

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

Marianne
Level 6
Partner    VIP    Accredited Certified
Seems @starra has not been back since Sunday to look for answers...

Found out that there's something wrong on the media server.

So i tried to use other media server and it works.! 

To answer your question, the retore was not writing for more than an hour. so i asked network team and they found out the media server is unstable interms of ingoing or out going of the data. btw they used ping test using the highest possible data packets.