Doug!
Just curious why you Mac guys are buying backup systems, when there is
a perfectly good way to use rsync. Here's my nightly backup script,
which currently sends my nightly incrementals to a cheap 3TB USB3
external drive:
I'ts the glitzy interface! Have you ever *SEEN* it? You feel like you
are traveling back in time and can find any previous state a the click
of a button! ;^)
Actually, there is *something* to this. In my case, it was after a
particular ugly crash/loss (you know how hard I can be on my equipment)
without a good backup... I hadn't *bothered* to set up and test a good
backup system and finally got caught with my pants tangled around my
ankles and decided that the incremental cost of a "Time Machine" over a
basic disk of similar magnitude (might have been $400 vs $300) would pay
for itself the first hour I *didn't* have to think too hard about how to
use it. I spent more than a little time convincing myself the damned
system actually *worked* but it wasn't until the first real emergency
use that I got much confidence.
I use even more clunky methods with my Linux systems and don't bother at
ALL with my Winderz versions... except maybe to make entire disk
images... but the fact is, these are not my *working* systems... I build
them and use them to test for *very specific purposes* and am as likely
as not to wipe and rebuild anyway for each new project.
I occasionally use TM (mostly for my wife) to simply go find an event
where she deleted or overwrote something she needed. Usually I can
find in e-mail the date/time of the triggering event, usually several
days to a few months previous, and then go bumble around in the Time
Machine until I *see* the (usually a flurry of) changes and forensically
can figure out exactly *what* changed and *guess* why, etc. A point and
click later and we are back to the earlier state, and if I'm wrong,
another point and click and we are at another state, and ....
Is there an equivalent process you use for recovering things you aren't
sure when/where/how you lost? Seems likely, but it is probably a little
harder to learn. I no longer have the benefit of working amongst a
group of like-minded Unix-Heads (though this list is a good resource)
such that this kind of folk-knowledge suffuses through my bony skull.
I also keep the state of my *entire* system backed up because (blame it
on Apple, but I think the package managers for Linux create the same
challenge) I have *no clue* where much of the state of my system is
actually maintained... when I install software, libraries, frameworks,
etc... I have at best a vague idea where it actually resides. To wit,
I *know* that most Mac Apps are this hairball of code and data and
libraries and pointers to dynamic libraries and simply dumping the
Application File somewhere (usually the Applications Directory) is all
that is involved in an install. Unfortunately, frameworks in particular
are a bit trickier... and I *do* sometimes wish I understood better how
all that worked (as I would *almost* have to if working closer to the OS
as with Linux), but frankly I'm too lazy... I'm usually trying something
out, and don't want to invest much more than a few point, clicks, and
"agrees" to give it a test drive. This means my system is pretty crufty
with installed stuff I never use. But so is my toolbench/toolbox....
tools I needed for *one job* still being carried around for no
particular reason!
Maybe I *will* use your method on my Linux systems... seems prudent and
easy enough. But then surely more than my *home* directory which is the
*least* of my worries on my Linux systems.
thanks!
- Steve
#!/bin/bash
# Just in case they are not mounted
/bin/mount /mnt/3TB >&/dev/null
/bin/mount /mnt/Movies >&/dev/null
/bin/mount /mnt/Video >&/dev/null
#
#/home/roberts
#
echo "Starting /home/roberts backup" >>/home/roberts/backup2.log
date >>/home/roberts/backup2.log
/usr/bin/rsync -vurltD --exclude-from=/home/roberts/.rsync/exclude
/home/roberts /mnt/3TB >>/home/roberts/backup2.log 2>&1
echo "Completed /home/roberts backup" >>/home/roberts/backup2.log
date >>/home/roberts/backup2.log
On Mon, Apr 8, 2013 at 12:02 PM, Steve Smith <sasm...@swcp.com
<mailto:sasm...@swcp.com>> wrote:
My $.02 on Time Machine.
I bought a 2TB time machine about 4? years ago and set up two MB
Pro's with it. Other than a little irritation from accidental
reboots on the device (connected to the same power strip as my
flakey motorola internet service, yielding a reboot via powerstrip
toggle sometimes), I've had nothing but good look.
I've only had one occasion to do a full restore in an emergency
and it worked like a charm.. I *have* used it to migrate between
MB Pros and an iMac about 5 or 6 times in the same period. That
has worked flawlessly as well.
It might be prudent to back that up somewhere offsite, but I'm
just not that prudent and now am spoiled to my regular "backup"
and potential "restores" being almost entirely invisible to me. I
can't tell from the discussion on the list how "transparent" the
true cloud services are, unfortunately I'm pretty sure my totally
lame internet would make *restore* a long and painful experience.
- Steve
I have one data point. One of our Macs near Seattle had a drive
fail, so I had an employee take it to an Apple store. The
'genius' was very happy when he saw the Time Machine, and, I
think, nothing was lost.
About the depth of cloud backups: I now use Arq on the Mac. The
backups are in Amazon's S3, and the frequency is settable: I have
one done every hour. You set a limit on how much space you want
to use -- just as a Time Machine has a fixed size -- and once you
hit that limit, it will overwrite the oldest versions as
necessary. Also the paid version of DropBox keeps at least some
history. For saving a Time Machine offsite, Amazons Glacier
storage is one cent a gigabyte per month, so your 150 gigabytes
would be $18 per year. They really hit you with transfer charges
if you try to read a large amount in a short time, but since that
presumably happens only when your Mac and your time machine have
both been roasted in a fire, you probably will be happy to pay
them. Unfortunately 150 gigs is not enough for most time machines.
--Barry
On Apr 6, 2013, at 8:42 AM, "Robert J. Cordingley"
<rob...@cirrillian.com <mailto:rob...@cirrillian.com>> wrote:
So has anyone successfully restored an entire system from the
Cloud (or a Time Machine come to think of it)? How easy was it?
Any statistics on success rate?
============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribehttp://redfish.com/mailman/listinfo/friam_redfish.com
============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
--
/Doug Roberts
d...@parrot-farm.net <mailto:d...@parrot-farm.net>/
/http://parrot-farm.net/Second-Cousins/
/
505-455-7333 - Office
505-672-8213 - Mobile/
============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
