On Tue, May 25, 2010 at 03:07:13PM -0500, Les Mikesell wrote:
> On 5/25/2010 2:49 PM, Robin Lee Powell wrote:
> >
> >>
> >> Without looking at the code, I'd guess that it would go through
> >> the rest of the list first before retrying failed jobs - but
> >> that's just a guess. Maybe it would help to lower
> >> $Conf{MaxBackups} if you haven't already.
> >
> > No can do; I have many other hosts that need to be backed up
> > besides this one.
> 
> Have you timed lower settings?  With most hardware it is probably
> counterproductive to run more than 2 concurrently on a LAN.  I've
> always thought it would be nice if you could specify groups with
> their own concurrency limit to accommodate multiple low-bandwidth
> links or alias that point to the same host/disk.

This isn't exactly over a LAN.

We have over one hundred hosts.  Many of them routinely take 4+
hours to backup.  Several of them take 12+ hours to backup.  One of
them takes more than a day.  Despite all that, and ignoring backups
on the semaphored host, only 3 backups are more than 1 day old right
now.

So, whatever we're doing, it's working just fine except for this
case.  Dropping it to 2 concurrently would be ... bad.

Our bottleneck is usually I/O on the remote server, FWIW.  Our
customers have some very unfortunate directory structers, and
GFS/RedHat Clustering really doesn't like them.

-Robin

-- 
http://singinst.org/ :  Our last, best hope for a fantastic future.
Lojban (http://www.lojban.org/): The language in which "this parrot
is dead" is "ti poi spitaki cu morsi", but "this sentence is false"
is "na nei".   My personal page: http://www.digitalkingdom.org/rlp/

------------------------------------------------------------------------------

_______________________________________________
BackupPC-users mailing list
BackupPC-users AT lists.sourceforge DOT net
List:    https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:    http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/