Menu
▾
▴
Re: [Bacula-users] Getting multiple streams for a large file system
|
From: John D. <dre...@gm...> - 2012-06-12 19:48:26
|
On Tue, Jun 12, 2012 at 3:31 PM, Clark, Patricia A. <cl...@or...> wrote:
> I have several large file systems (>1TB) where I want to break them up to get smaller backup streams in parallel to increase the throughput to tape. My fileset directive is below. I want everything in /home, but divided by the regular expressions provided in the Options section. Will this do what I want or will I only get one backup stream?
>
> FileSet {
> Name = "homedirs2 Set"
> Include {
> Options {
> signature = MD5
> RegexDir = "^/home/[0-9]*"
> RegexDir = "^/home/[a-e]*"
> RegexDir = "^/home/[f-j]*"
> RegexDir = "^/home/[k-o]*"
> RegexDir = "^/home/[p-t]*"
> RegexDir = "^/home/[u-z]*"
> RegexDir = "^/home/[A-Z]*"
> }
> File = /home
> }
> #
> Exclude {
> File = /proc
> File = /tmp
> File = /.journal
> File = /.fsck
> File = lost+found
> }
> }
>
You want to run X jobs with X different filesets. The config above
will certainly not accomplish that. It will just be 1 job and 1
fileset. Also you will need to enable concurrency and have multiple
storage devices since a single storage device can load only 1 volume
at a time. Think of each disk storage devices as a single tape drive
with all the abilities and restrictions.
John
|
