cancel
Showing results for 
Search instead for 
Did you mean: 

Restore is failing for Linux Client with 2826 and 41 error code

Stiva
Level 4

Hi All,

Backup is failing with error code 2826 and 41. I have increased the timeout limit to 3600S but still Restore is failing.

It seems that restore is going in hung state for some time and fail with Timeout error.

master server windows 2003 and netbackup version 7.1

client server Linux and netbackup version 7.1

Logs details:

9/26/2014 4:19:55 PM - Info bpbrm(pid=15052) child done, status 41        
9/26/2014 4:19:55 PM - Info bpbrm(pid=15052) sending message to media manager: STOP RESTORE auvxxxxxxx_1411550499    
9/26/2014 4:19:56 PM - Error bptm(pid=732) The following files/folders were not restored:      
9/26/2014 4:19:56 PM - Error bptm(pid=732) UTF - /xxxxxxxxxxxx         
9/26/2014 4:19:56 PM - Error bptm(pid=732) UTF -  /xxxxxxxxxxxx         
9/26/2014 4:19:56 PM - Error bptm(pid=732) UTF -  /xxxxxxxxxxxx          
9/26/2014 4:19:56 PM - Error bptm(pid=732) UTF -  /xxxxxxxxxxxx          
      
9/26/2014 4:19:56 PM - Error bptm(pid=732) more than 10 files were not restored, remaining ones are shown in the progress log.
9/26/2014 4:19:56 PM - Info bpbrm(pid=15052) media manager for backup id auvxxxxxx_1411550499 exited with status 150: termination requested by administrator
9/26/2014 4:19:56 PM - restored image auvxxxxxx_1411550499 - (network connection timed out(41)); restore time 01:04:15
9/26/2014 4:19:56 PM - Warning bprd(pid=6812) Restore must be resumed prior to first image expiration on 10/25/2014 8:21:39 PM
9/26/2014 4:19:57 PM - end Restore; elapsed time: 01:05:00
invalid error number(2826)

Thanks in Advance!!

1 ACCEPTED SOLUTION

Accepted Solutions

Stiva
Level 4
After observing the tar logs properly we got to know that the restores were failing due to network connection was broken. Possibly because it was using the default net buffer size i.e. 64KB. NET_BUFFER_SZ is a file containing a number indicating the TCP/IP socket buffer size that should be used for data transfers between the NetBackup media server and its clients. Most networks can handle a NET_BUFFER_SZ of 262144 (256K). So we increased the value of NET_BUFFER_SZ to 262144 on both Master and Client server. It helped us to increase client read time and restore performance. The issue has been resolved and restore completed successfully.

View solution in original post

5 REPLIES 5

RamNagalla
Moderator
Moderator
Partner    VIP    Certified

does it failing immediatly the job start or after reading/writing some data ?

apporxmatly how much of data you are trying to restore?

 

did you increase the timeout value for media server of client?

 

if the restore is triggred from Mater server, attache the restore log this post

default restore log location /usr/openv/netbackup/logs/user_ops/<user_id>/logs

user_id is the one the ID which is used to trigger the restore...

Stiva
Level 4

No It is not faling immediately It writes some data then it will fail. Apprximately 40GB trying to restore. Media and Master server is same and timeout limit is 3600Sec.

Please find attachment for logs.

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified

So, your second attempt (resume) was successful.

62 out of 63 files were restored, 1 file was kept because you chose not to overwrite existing files.

If you are still experiencing issues, ensure all of these log folders exist before you start another restore:

On media server: bptm and bpbrm

On client: bpcd and tar

Stiva
Level 4

Hi Marianne,

Still I am facing same issue. Please find attached logs.

Stiva
Level 4
After observing the tar logs properly we got to know that the restores were failing due to network connection was broken. Possibly because it was using the default net buffer size i.e. 64KB. NET_BUFFER_SZ is a file containing a number indicating the TCP/IP socket buffer size that should be used for data transfers between the NetBackup media server and its clients. Most networks can handle a NET_BUFFER_SZ of 262144 (256K). So we increased the value of NET_BUFFER_SZ to 262144 on both Master and Client server. It helped us to increase client read time and restore performance. The issue has been resolved and restore completed successfully.