Hi all,
long time since I've been in so, Hello!

after a disastrous couple of hard drive failures I've ended up in the situation where all of my 2-300gb of mp3 is spread across about 4 players,

I know I've taken music from players to the pc before but in emplode I can't see how it's done, copy / paste in any form seems out of the question, I'm having 'one of those mornings' so should probably leave it until I'm thinking clearly but..

any idea's gratefully received!

thanks,
Hugh

Thank you Tony - I just couldn't remember how to do it!! Luckily I've never really messed with the id tags - empegs are wasted on me really but, love em!

only 28hrs to go on this 30gb player, then there's the 10, 60 and -omg.. the 150 that's full!

I'm guessing it would download a lot faster if connected to a network?

argh... left it downloading overnight, only three hours to go but, instead of downloading as folders, it's just downloading every individual track into one folder... why oh why didn't I check last night before going home!

Yes, Ethernet is about twice as fast as USB.

http://riocar.org/modules.php?op=modload...;faqent=192#192

If you use Jemplode instead of Emplode, there are extensive options to control folder structure, file names, and tag write backs when you do a download from the player.

thanks Tony, I think I used jemplode before - now you mention it. probably forgot to re-install it after the hdd failures.

Hey guys,
what is it with hard drives these days? they barely last a year or so... my 1TB back-up HDD had just died, along with my music files which appear irretrievable by my local pc place...

yet again I need to repeat the aforementioned proceedures to take the music from my players & rebuild my music library, I've downloaded jemplode but it won't locate my player(s) via USB, Emplode has found it no problem.

Help appreciated as I can't find / make out any answer in the faqs.

thanks in advance.
Hugh

a bit long winded but I could make a new file tree in my music folder and simply download each playlist into it's respective duplicate folder.

however, it takes forever to download tracks (maybe 30 seconds on average).. I just transferred 30gb of mp3 files and folders (via usb) from a back up 2.5" mobile sata disc drive in less than 15 mins.. would it be quicker for me to put each drive into the back-up cage and simply rip the folders away directly - or does it not work like a memory stick / back up folder??

sorry for the stupid questions. I've recently moved to a new business unit and don't have a network set up yet, it's all usb I'm afraid.

Jemplode only works via ethernet, if I recall correctly. Connect to the player with ethernet to retrieve your files.



By definition, the drive in question was not a back-up drive, since it contained the only copy of your music files. I would call that an external or secondary drive, specifically one that has not been backed up.




Agreed that hard disks die a lot. I recommend backing them up.



That depends on the brand you buy, and how you care for the drive.

Recently the drive in my laptop died because I dropped something small on the laptop at exactly the right spot at exactly the right time, enough for it to crash the heads. (Luckily I had it backed up.) Other things that can make a drive die are letting it get too hot or too cold. Or they can Just Die if they're Seagates.

I spent some extra money and got a Solid State Drive for my laptop, in hopes that it will be more reliable. But I still back it up, and the most critical data I back up to multiple places.

That's because you're using USB. It's slower than ethernet.



If you want it faster you have to graduate out of USB.

If you don't have a network, most of the time you can simply plug an ethernet cable directly from the computer to the empeg.

Assuming that both devices are configured for automatic DHCP addressing, then both devices will get a UPNP address that will look like 169.254.x.x, and Emplode and/or Jemplode should be able to find the player.

thank you Tony.. I'll try the ethernet cable idea in the morning when I get back in to the office.

re the back-up.. agreed, by definition it's not a back up as it was the only copy.. technically it wasn't the only copy until a hdd in the pc died a few weeks ago when I moved offices!! I hadn't used the back up in 6 months or more and it just doesn't want to fire up properly..

H

Or any other brand you care to mention. Everyone has horror stories with different brands.

You should have at least three copies of data you care about: (1) the one you're using; (2) a backup; (3) another backup, stored somewhere else. You should rotate these periodically. And remember: it's not a backup process unless you regularly test the restore process. For my Windows Server box, I use BackupAssist, with "Daily 1", "Daily 2", "Weekly 1" and "Weekly 2" disks in rotation. The "Weekly 1" and "Weekly 2" disks alternately live in a cupboard in the Electric Imp office in Cambridge. I don't test the restore as often as I should, though.

There's nothing on my NAS box that I particularly care about, though it would be a PITA to re-rip all of my DVDs and BDs, so I am considering setting it up to back up to Amazon Glacier. Everything that I care about on my Linux box is in GitHub...

Agreed that any hard drive can die at any time, regardless of brand, and thus the importance of backups.

And I also see that the original Backblaze article that I linked, showing statistical support for my personal experience that Seagate sucks, has been criticized for its methodology. So it's not as simple as "Only Buy Hitachi".

ironically it's a hitachi that's just died on this occasion.. a brand I've always trusted!

Proving Roger's point.

Now I'm wondering if there's an easy way to set both a Time Machine remote disk (a Time Capsule, basically, just provided by a Mini and not Airport hardware) and a RAIDed disk on the same mini to both do sync'd backups to Glacier. I guess I should do some research

OS X Server has an option to host a Time Machine share, and it's doable without OS X server, just a little more manual. You can stash the share anywhere on the machine, included RAID disks if you wanted.

Machines backing up to it will be creating a sparseimage. As long as they aren't encrypted, or you are okay with keeping the password on the server, you can then mount those sparse images and pull the data from the /latest . Issue is, when the sparseimage is mounted to mirror to glacier, clients may get errors. Including possibly a dialog that would wipe the sparseimage and start over.

You could possibly mount it in the middle of the night, mirror it locally, then sync that mirror to glacier. Still could cause some errors on Macs due to time machine backups attempts during a powernap.

Or possibly host the sparseimages on a volume that allows snapshots, and just mount a snapshot instead.

I've been meaning to do something similar. Until then, I just ended up throwing Backblaze at the situation.

That's all right for you guys with a fast upload connection, but in my case it would take 111.6 days*, 24 hours a day, to back up my 2.3TB of data at the 2mbps maximum upload speed I have available. And it isn't always that fast. Figure four months to get it all uploaded, then add another month or so onto that for the additional data that gets sent up every time I change or move a file. For example, every time I load an audio book into my iPod, the file gets moved from the "Unheard" directory to the "Heard" directory. That's half a gigabyte to delete and back up again, unless Backblaze is smart enough to realize that the data hasn't changed, only the directory tree. (My backup software isn't that smart, it deletes and re-backs it.)

So, I back up to an internal drive. Then I back up to an external drive. Then I back up to an external drive that is kept off-premises. My most valuable data I keep in a backup directory on the original drive, as well as on the backup drive and the off-premises drive, giving me a total of six copies of the data including the original.

Hey, just because I'm paranoid doesn't mean they're not out to get me!

tanstaafl.

*2.3TB x 1024 = 2355GB x 1024 = 2,411,725MB x 8 = 19,293,798Mb divided by 2mbps = 9,646,899 seconds divided by 3600 = 2,680 hours divided by 24 = 111.6 days.

Such technology can save a lot of people upload time too. Think of all the common files we all have on our machines. The same version of a browser, or the same music downloaded from SXSW. Only one person has to upload it, everyone else simply gets the reference added to their account. Or for my two machines, duplicate personal files between them only crossed the wire once.

Crashplan is smart enough to deduplicate so if you are just moving files around, they won't be resent. Even if they are changed slightly only the differences need to be sent.

Not sure on BackBlaze whether it does parts of files or just whole ones, but the technology absolutely exists.

As for 111 days I think my initial backup was about 100 days but I just let it chug along, prioritising certain folders first. Once that was done I added more less important stuff and just let it chug along in the "background". It got there eventually.

I'm a big believer in automation of backups. If you have the resolve to manually do your rotation etc, great.

We don't all have fast internet connections. My current one is 2-3mbps up and my previous one was only 0.8mbps up.

Despite that I have close to 2TB backed up with CrashPlan. Sure, it took a while to get everything backed up initially.

But the way CrashPlan works makes it easy to prioritise different files. So for example my documents and source code have the highest priority, then my photos, then my ripped FLAC files and finally my massive virtual machine images.

So when I started using CP all my documents and source code were backed up in the first day or so, then it did my photos and then the rest. And the same priorities still apply, so if I dump a whole load of new photos and FLAC files onto the machine it will do the photos first and then the FLAC files.

CP also lets you backup between machines, at the same time as backing up to their cloud storage. So my laptop is backed up to my home server, while also being backed up to the CP cloud. And the server backs all its data to the CP cloud.

I also have CP running in some of the virtual machines I use for my day-to-day development work, so any code I'm working on there gets backed up on the fly all through the day.

I also throw DropBox into the mix, so that most of my data lives in a good half dozen places. One has to be careful with DropBox though, it is very good at distributing your file management mistakes... (my entire DropBox contents is backed up to two places by CP)

I admire people like who can rigorously apply a manual backup strategy, but it would never work for me. For lazy people like me the backup has to be completely automatic.

It works for me only because none of my data is mission critical. If I lost everything I had done since my last manual backup (about four or five days ago) I'd say, "Oh, darn, that's too bad.", and I'd go about replicating what I could of it (probably about 75%), and then I'd not lose any sleep over the rest of it.

tanstaafl.

No, it works for you because you actually do th backups. If I was in your place I'd do them for a couple of months and then slowly stop.

So I won't have a backup from four or five days to go back to

Would you really not be bothered if you lost all your photos ?

I won't lose my photos, because they are on my D: drive, which is backed up to my E: drive, which is backed up to my L: external drive, which is backed up to my M: off-premises drive. The most I could lose would be four or five days of most recent photos.

tanstaafl.

And you do this more than once a week?!

-jk

Sometimes once a week or so, other times every couple of weeks, it depends on how much new data has been added. Remember, none of my data is mission critical, if I lost the most recent week of stuff, it would be a minor annoyance, not a big problem. Most of what I lost (health insurance claims, spreadsheet updates, etc.) could be re-created if necessary in any case.

tanstaafl.

So here's a piccy of the 9 Samsung 1Tb drives I bought

Oh, and one or two more at the peak of the RMA failure cycle...