Hopelessly confused with DEVONthink Sync

DEVONthink Feedback
1

I’ve read many highly informative threads on these forums on how to go about syncing a Devonthink database between two computers, and with each one I read I become even more confused than I was before. I’ve also watched the tutorial video on syncing, and the same is true.

Current Problem & Desired Outcome
I have a 20GB DTPO database stored in my Dropbox folder so that I can access it from my MacBook and my iMac, and I have been using Dropbox to keep this database in sync between the two machines (making sure not to have the database open on more than one machine at the same time). I KNOW THAT THIS IS HIGHLY FROWNED UPON AND IS A RECIPE FOR DISASTER, so once and for all I’m trying to figure out how to go about setting things up so that I can sync “properly” using Devonthink’s built-in sync capabilities.

Where My Confusion Lies
I am unclear as to which sync method (or methods?) I should be using.
• Should I authorize my Dropbox account in the sync preferences of DTPO on both my iMac and MacBook computers, and then specify that I want to sync my database using Dropbox?
• Do I add a new Local Sync Store and save it to ~/Dropbox/Apps/DEVONthink/.
• Are sync methods meant to be used in combination with one another (such as setting up a Direct Connection sync to speed up syncing my between my laptop and desktop machines while they’re both on the same local network.

Misc Bullet Points:
• I understand that the actual Database.dtBase2 file needs to be stored outside of my Dropbox folder… that much I’ve got.

• I’d like to be able to sync my database in/out of my local network.

• I’ve got 100GB of Dropbox strorage space, so keeping my 20GB Devonthink database in Dropbox isn’t an issue.

• If possible, I’d like to avoid having to re-upload/download my existing 20GB database to Dropbox’s servers, which would take several days. If it somehow simplifies the process if I do so, that’s fine I can just suck it up, but I’d rather avoid going there if possible.

• Worth noting… when I say I’d like to 'Sync my Database" what I mean is that I’d like to be able to fire up Devonthink on either machine and know that any file I added to the database on one computer will show up when accessing that database on the other computer, and vice versa.

If you’ve made it this far, thanks so much for your time! I can’t wait to hear back from you guys with any helpful guidance you have to offer. Thanks!

**Please note… if I am wrong about any of this, please don’t hesitate to let me know, but please avoid extraneous details unless they help me accomplish my goal outlined above. Too much noise makes me want to jump ship… (by step #20 of this explanation ~ Syncing and database location/storage ~ my brain nearly exploded).

2

Bueller?.. Bueller?

Any help would be sincerely appreciated! Thanks!

3

The simplest steps (assuming the database is in Sync on both machines, or it doesn’t matter if it is or not)…

  1. If a copy exists on Machine B, select File > Export > Database Archive to save a zipped version of the database. Note: You could also do a Select All on the database contents and do a File > Export > Files and Folders… to a newly created folder on your Desktop. The point is: You don’t want a copy on Machine B when you start this process.

  2. Right-click on the database in the Global Sidebar and choose Delete Database. (You’ve got a zipped copy as a backup.)

  3. On both machines, set up a Dropbox Sync location in Preferences > Sync, and YES you want to tell Dropbox it’s okay to allow this.

  4. In the Dropbox application’s Preferences > Advanced > Selective Sync, uncheck the /Apps/DEVONthink folder. (You will not need to use the Dropbox app for DEVONsyncing.)

  5. Sync Machine A pushing to Dropbox. Wait until it’s finished. With a large database this could be a considerable amount of time.

  6. On Machine B, select Import Database… from Preferences > Sync and follow the dialogs.

  7. At one point you will be asked to save the database to Machine B. Put it somewhere like your User Documents folder but not your Dropbox folder.

  8. Done.

• The same essential process applies to WebDAV and local syncStores as well. Direct Connection requires you to not have a local copy on Machine B too but there’s no intermediate sync location.

•• Note that Syncing changes to Dropbox does not update the other machine. Changes are applied when the other machine Syncs.

••• It is always a good idea to do a Tools > Rebuild Database before starting this process.
••• It’s also a good idea to periodically do a Tools > Verify & Repair to make sure your database is coherent and intact. (You get your oil changed occasionally, don’t you? 8) )

•••• Indexed files will Sync along with imported data. This is by design (and after a lot of discussion with Users while Sync was being developed).

(PS: Don’t forget, we have a lot going on behind the scenes here. Just a reminder.)

1 Like
4

Thanks for your response (and sorry for my impatience).

I tried something similar to the procedure you outline, but I stopped it when I noticed that it had added what looked like a 19GB copy (or sync-able version) of my Database in the ~/Dropbox/Apps/DEVONthink/ directory, and then it began to upload the 19GB worth of new sync-able data.

Couple of quick questions so I can have a better understanding of what’s going on here:
• Assuming I do proceed with this approach, do I need to just deal with the fact that I will now have two of these roughly 20 GB database files taking up space on my laptop and desktop hard drives (40GB of space on each computer? Really???).
• Am I able to delete the initial database once the sync-able version is safely created in Dropbox and is finished being uploaded? (I assume not, but I gotta ask).
• With Dropbox set to selectively sync and basically ignore the copy in ~/Dropbox/Apps/DEVONthink/, then what’s doing the actual copying/syncing here if not Dropbox?
• Is there even a copy of this database file stored on Dropbox’s servers somewhere that I could restore in the event of a catastrophe (like both my computers spontaneously combusting at the same time)?

Thanks again for helping me out with this!

5

No, it’s not 40GB because it is not Syncing back to your Dropbox folder. (Part of the reason you exclude the Apps/DEVONthink folder.) Your data does not reside solely on Dropbox’s servers. You have a local copy (just like your DEVONthink database lives locally on your drive) and a remote copy of the syncData on Dropbox’s servers.

Absolutely not. It is your live database on your local machines. You are not working off the remote files.

All transactions between DEVONthink and Dropbox are done via Dropbox’s APIs. It has nothing to do with the Dropbox application at all.

The syncData is a copy of the database, just not one as you’d see in the Finder. It is storing the data required to reconstruct the database or it’s changes on another machine running DEVONthink.

6

But once it’s all said and done, I’ll have 20GB less hard drive space on my primary internal hard drive, because not only will 20GB be taken up by the database stored in ~/Documents, but an additional 20GB will be taken up by the (sync-able) database stored in ~/Dropbox/Apps/DEVONthink/ folder. Is this correct?

7

No, as Jim said, you will have your database file stored on your computer, and the Sync data stored in the cloud, on a Dropbox server. You did exclude the DEVONthink data folder from your Dropbox folder, right?

8

I haven’t done it yet, but I will follow all the outlined steps now and I’ll let you know how I make out.

Thank you guys for your help! I’ll keep you posted.

9

Absolutely no luck getting Dropbox sync working!

First I ran into this issue having something to do with ‘Restricted Content’ (Problem Syncing Omnioutliner files), then I followed the steps offered by the DT user in that thread by using the ‘Clean Location’ option, but I subsequently wound up experiencing another Dropbox sync failure because of an error citing ‘Missing Content’ (because I’d previously tried removing the restricted content after I got the first error message).

Finally I realized that none of this was worth it! Jesus what a headache. I don’t remember the last time a piece of software gave me this much trouble. I’ve tried several times over the last couple years to configure syncing the way you guys suggest configuring it, and every time, for one reason or another, I always wind up quitting because it’s too much of a hassle.

At this point, I’m now going to resort to doing it how I was doing it before, which worked… using Dropbox to directly sync the databases, making sure not to have the same database open on two computers simultaneously, and making sure to keep versioned Time Machine & Chronosync backups of my database in the event of disaster.

Let’s hope that whatever you guys eluded to having “in the pipeline” has something to do with the sync capabilities, because right now its terrible!

I love Devonthink, but it has two major shortcomings that are virtual deal-breakers:
• Syncing - Make it’s sync capabilities work as seamlessly as Evernote and Dropbox sync their data
• Mobile Access - allow users access to their entire databases from their iPhone/iPad without the need to manually sync bites and pieces of their databases using the Mobile Sync half-measure (e.g. make it work more like Plex, AIrVideo, or Stream To Me… how they allow you to stream movies from your home computer through the use of a media server that runs on the computer at all times… I’d have no problem running a Devonthink Media Server on my primary computer 24/7 if it meant I’d have access to all of my files from my mobile devices).

Thank you for your attempts to help.

10

Just for the record, the “Restricted Content” issue is a bug with Dropbox, not with Sync. This issue was never encountered until a couple weeks ago, when suddenly several users reported it at once. The Dropbox-related code in Sync hasn’t changed for over a year at this point. If you view the records in question through Dropbox’s web interface, you get the same error. QED.

11

I recently signed up for the free Box 50gb storage.

I now keep all my databases in Box (they are under 50gb)

I don’t using the sync. I just keep the databases in Box and it automatically syncs it, but there is a local copy.

This seems to work for me…but the only thing the you need to remember is to close Devonthink. You should only have one copy working when it syncs.

12

If this is Syncing like Dropbox, this is no more safe a practice. Personally, I would NOT do this.

13

Have not completed this yet because I need to get machines a (desktop) and B (laptop) in the same room, on the same network.

My main concern is that if I have two databases with the same name on each machine, won’t the database on machine B go completely missing, since as far as I can tell it was deleted. (But backed up on the desktop?

Seems like I am missing a step somehow.

Thanks for your time.

14

My main concern is that if I have two databases with the same name on each machine, won’t the database on machine B go completely missing, since as far as I can tell it was deleted.

I’m not sure what you mean here.

If you had two databases with the same filename (which I would not advise you do), say DB-A, they’d have to be in separate folders since the filesystem won’t allow files with the same name.

If you imported a database from the sync location in DEVONthink’s Preferences > Sync…

  • If the database had already been imported to the machine, DEVONthink would identify it by its UUID and open it.
  • If the database hadn’t been imported - and DEVONthink would know this because of the UUID - and you saved it to a location where you had a database named DB-A, then yes, you could overwrite it. However, this would also be an unusual circumstance and one I hope you’re not in.
15

ok, seems like I’ve made big mistake here, as I made some big assumptions. I will try to describe below.

Say I have a database on machine A called Admin (I call this database admin.a)

I also have a database on machine B called Admin (I call this database admin.b)

As described they are both called admin on the different machines - for reasons and assumptions that will be described below.

I have let them both the databased on machine a and machine b (admin.a and admin.b) grow on each machine independently (as previously stated, each of the databases on each different machine are called Admin). I assumed that one day they will both live in a database on a server (say a dropbox server) in a database that somehow combines admin.a and admin.b (call it admin.s - s for server). I initially didn’t do this because I thought with a dropbox account it would be ultimately be easy, and that I could do it any time, and that is might just be a matter of learning to check a box or two.

I assumed that when admin.s is first developed on the server (dropbox say), it would initially throw away any conflicting versions of the files from both admin.a and admin.b (maybe pick the one with the most recent date in this case) and otherwise combine the files from each admin.a and admin.b in admin.s (on the dropbox server). It would then push admin.s (the new combined database that lives on the server ) back to machine a and machine b, as the new admin.a on machine a and admin.b on machine b.

“My method” says as described would eliminate old duplicate files in both admin.a and admin.b, and this clearly would be undesirable for some people. It will not destroy entire folders that exist on admin.a and not on admin.b and vice versa however.

However there doesn’t seem to be a way in the steps provided in the example to pull all of the information from database B back into the process. It was zipped and left on the desktop of machine B, but all of the data included in this database doesn’t seem to be reintroduced into database.b or into database.s. All the information just from database b just seems to be lost in the process described.

The way you described the sync makes sense. However it turns out my initial assumption may have been costly, as I lost access to all the data in admin.b

Is there a way to introduce the information from the initial admin.b back into the process?

Has the information already been reintroduced and somehow I just missed it?

I know this is long. Hope this makes some sense. I edited it a bunch of times, which in my experience can add more confusion and typos than it removes.

Thanks for your time.