(like not scheduling large
backups to go on tape first, so the tape gets filled maximally; or not
first backing up hosts that haven't been backed up for a while, so they're
more likely to get to tape instead of continuing to fail;
Give it some holding disk space and these things take care of themselves
as you flush anything that didn't fit on the first try. You shouldn't
ever have anything that continues to fail (or even fails once).
when I say 'fail' here, I mean "amanda won't do a full backup, because it
thinks its out of space" (i.e. I have less space on tape than the data I
want to back up; and no money for a changer). Eventually the last full
backup gets overwritten, and you have (effectively) a 'failed' backup. See
below.
or easily telling
it "back up as much as you can to tape, leave the leftovers on disk to be
flushed manually").
If that doesn't happen by itself you have something configured wrong.
Your 'reserve' value has to be reasonable
I eventually got this right; but only after someone explained it to me on
the amanda mailing list... the documentation in the amanda.conf file is
completely counter-intuitive:
# If amanda cannot find a tape on which to store backups, it will run
# as many backups as it can to the holding disks. In order to save
# space for unattended backups, by default, amanda will only perform
# incremental backups in this case, i.e., it will reserve 100% of the
# holding disk space for the so-called degraded mode backups.
# However, if you specify a different value for the `reserve'
# parameter, amanda will not degrade backups if they will fit in the
# non-reserved portion of the holding disk.
reserve 90 # percent
<<<
it makes it sound like the 'reserve' value is how much you're saving for
_degraded mode_ backups. If you squint your eyes and read it enough times,
you might come up with the correct interpretation (that 'reserve' is how
much you save for non-degraded-mode); but it's not what you'd read from a
first glance.
even with this set; it *still* puts itself in degraded mode when it runs out
of space on tape (according to what the specified size of the tape is);
refusing to ever do a proper backup of some machines. what I've ended up
doing is lying about the tape size; so that it hopefully does as many normal
backups as possible, before running out of space on tape, and then going to
degraded mode (even tho there might be 30-40GB of holding disk space left to
use for normal backups).
in the end, lengthening the tape cycle solved most of the problems; letting
me space out full backups farther in between the incrementals, so the amount
it's trying to back up at once is less.
Meanwhile amanda does the best job I've seen at making
sure that within every set of 10 tapes there is at least one full run
of every filesystem, and every night it has at least an incremental
of each one and the incremental levels are the lowest that are likely
to fit. I don't see any way to get those concepts to mesh with
backuppc.
amanda does a decent job for the environment it was designed for... a
University, back when tape was big and disk was small. many machines backed
up to one tape. this is no longer the world we live in... disk is big, tape
is small; disk is cheap, tape is expensive.
BackupPC wins by far in this world.
Carl Soderstrom.
--
Systems Administrator
Real-Time Enterprises
www.real-time.com
-------------------------------------------------------
This SF.Net email is sponsored by Sleepycat Software
Learn developer strategies Cisco, Motorola, Ericsson & Lucent use to deliver
higher performing products faster, at low TCO.
http://www.sleepycat.com/telcomwpreg.php?From=osdnemail3
_______________________________________________
BackupPC-users mailing list
BackupPC-users < at > lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/backuppc-users
http://backuppc.sourceforge.net/
