Ok I'm pushing and try to keep it to mailing list :-)

On lundi, 9 janvier 2017 12.31:13 h CET Robert N wrote:
> let me see if I get this right. do you meand I should be using spooling for
> the backup to disk jobs, and copy jobs from disk volumes to tapes without
> spooling?
> 

I was addressing only one type of backup, several jobs running concurrently 
spooled, then merge in one device (disk based) (one or splitted by size 
volumes).

Not playing enough with tapes, I can't say anything on the second part (copy/
migrate) to tapes.

But then you are reading one device and writing to another one, the slowest 
making the io bandwidth.

It seems from what is expressed on the mailing list, that copy/migrate with 
jobs are failing in a one after the other syndrom.
But I can't say more on that

> On Thu, Jan 5, 2017 at 9:42 AM, Bruno Friedmann <[email protected]>
> 
> wrote:
> > On mercredi, 4 janvier 2017 21.23:21 h CET you wrote:
> > > speed, if I use interleaved jobs on disks, copytotape takes very long...
> > 
> > Which is why there's spooling to addresss that. with big chunk of spool
> > (I'm
> > using 128 to 512 GB) most of the job are spooled in one time and then
> > written
> > done to the media, with no interleave.
> > 
> > 


-- 

Bruno Friedmann 
 Ioda-Net Sàrl www.ioda-net.ch
 Bareos Partner, openSUSE Member, fsfe fellowship
 GPG KEY : D5C9B751C4653227
 irc: tigerfoot

openSUSE Tumbleweed
Linux 4.9.0-2-default x86_64 GNU/Linux, nvidia: 375.26
Qt: 5.7.1, KDE Frameworks: 5.29.0, Plasma: 5.8.5, kmail2 5.4.0

-- 
You received this message because you are subscribed to the Google Groups 
"bareos-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to