cancel
Showing results for 
Search instead for 
Did you mean: 

backups jobs always getting cache hits: 0 (0.0%)

Juasiepo
Level 4

Hi there,

I am experiencing this problem with all backup jobs, no matter if they are using server deduplication or client-side deduplication

Dedup seems ok with very good ratios but cache hits is always 0.

Here you are an example:

01/17/2013 07:15:33 - Info bak002 (pid=19972) StorageServer=PureDisk:bak002; Report=PDDO Stats for (bak002): scanned: 153346694 KB, CR sent: 617422 KB, CR sent over FC: 0 KB, dedup: 99.6%, cache hits: 0 (0.0%)

Surely is a configuration problems as it happens in all my clients but till now I haven't found the cause

My environment:

  • 1 Master 7.5.0.4 Linux
  • 1 Media 7.5.0.4 Linux
  • Clients Windows/Solaris/linux 7.5.0.4

Any idea?

Best Regards

1 ACCEPTED SOLUTION

Accepted Solutions

Juasiepo
Level 4

Hi

Finally I've got a reply from symantec support:

  • This is the normal behaviour, with accelerator it is normal to get cache hits: 0 (0.0%)

  • The Accelerator is Better Technology to  use.

Best regards

 

PS: Thank you Simon, I read the FAQ but I saw nothing related to this issue

 

View solution in original post

3 REPLIES 3

Juasiepo
Level 4

Hi,

I have disablesd the accelerator option in one of the policies and now the cache hits is almost 99%.So it seems that this was the cause.

The question now is what is better: ¿Accelerator enabled or high cache hits ratio?

Can anyone point me in the right direcction?

Thank you for your time.

Best Regards

S_Williamson
Level 6

Not the exact answer you might be looking for, but here is a good read on using the Accellorator feature and exactly why its better to use.

 

http://www.symantec.com/connect/blogs/frequently-asked-questions-netbackup-accelerator

Simon

 

Juasiepo
Level 4

Hi

Finally I've got a reply from symantec support:

  • This is the normal behaviour, with accelerator it is normal to get cache hits: 0 (0.0%)

  • The Accelerator is Better Technology to  use.

Best regards

 

PS: Thank you Simon, I read the FAQ but I saw nothing related to this issue