Trouble replicating to Dropbox Folder


 I hope somebody might help me with this. In DTpro I have a Group called "Aviation". In that group, I have one individual pdf (no subgroup) and a subgroup called "SOPs", with 18 pdfs. I replicated the group "Aviation" to my indexed "Dropbox" group. I then right clicked the "Aviation" group and "move to external folder." Everything showed up in Dropbox as in DTpro except for the one individual pdf that did not reside in the subgroup. Why is this?
 As part of my workflow, I need to access some DT information away from home. I've started to use Dropbox for this. Any group in DT that I might need somewhere else, I replicate it to my indexed Dropbox folder, then "Move to External Folder". It seems like individual files not located in a subgroup do not move into my Dropbox. Any ideas?? Thanks!


Just an update as I’ve been somewhat frustrated with this. I think I’m missing something simple. I just started over and re-did the workflow. I replicated the entire “Aviation” group to my indexed “Dropbox” folder. Then right clicked the “Aviation” group and “move to external folder.” Now only the individual pdf moved to dropbox! The subgroup “SOP” did not. I have to be missing something here. Any help is appreciated or any ideas on this type of workflow would also be appreciated.


Well, this sounded curious to me as I index most of my documents and I have never seen this behavior before. I just tried your example, following it step-for-step and all documents and folders were exported to my Dropbox folder. Might there be something else about the individual PDF that is an issue? One think I can think of is if the individual PDF were already indexed, it will not be moved out to the Dropbox folder.

Another thing to try is to move the Aviation group to the indexed Dropbox group instead of replicating it there. Then do a move to external folder and see what happens.

Thanks Greg. I shut down everything. Made sure all my trashes were empty and re-did the whole workflow, this time duplicating the group. It finally worked. Maybe I just ran into a glitch somewhere.

Anybody else use a similar workflow, i.e. Using Dropbox to access documents from DT that you need elsewhere? I’m hoping to just be able to take any group from DT, duplicate or replicate to my indexed Dropbox group (it’s the only indexed folder I use), then have access to it anywhere. When I’m finished with the files I can delete them from Dropbox, but still have them in the original DT group. I’m curious as to others’ thoughts on this workflow.


Glad to see it is working now. I think you will find that some variation of your Dropbox workflow is pretty popular. I don’t use Dropbox much to access documents from DEVONthink (I use DEVONthink To Go for that) but I do index my text files from Dropbox into DEVONthink.

I have used this kind of workflow (replicating files and folders to an indexed dropbox group/folder) for more than a year and it works fine.

However, some of my friends and colleagues have difficulties with this workflow, not because of bugs but because they do not understand what’s happening when they replicate an entry in DevonThink, index a folder, or move a file to an external folder and also because they don’t understand the differences between a group and a folder and between a replicant and an alias.

To avoid problems I suggest the following rules:

• never replicate an entry that is already in an indexed group to an indexed group
• do not rename, move or delete items in the indexed folder
• do not alias items in an indexed folder to an indexed folder

Having only one indexed group is a good idea, because it diminishes the chance to get replicants within the indexed group. Nevertheless, be careful. For example if you have a file ‘dogs’ with replicants in your ‘carnivores’ group and in your ‘pets’ group and you replicate both the ‘carnivores’ group and the ‘pets’ group to your dropbox group you might get problems. For that reason, I tend to prefer an indexed dropbox folder/group with no subgroups/folders.

Deleting files from the indexed folder on Dropbox might give troubles. Suppose you have a file ‘dogs’ with replicants in both your dropbox group and a pets group outside your dropbox group and that this file is moved to the external folder. This means that the dogs entry in the ‘pets’ group and the dogs entry in the dropbox group refer to the dogs file in the dropbox folder. So when you delete the dogs file from the dropbox folder the dogs entry in both the dropbox group and the dogs group refer to a file that doesn’t exist any more. The entry in the dropbox group will be removed when you update the index, but the entry in the dogs group will remain were it is, but it doesn’t refer to anything.

It is safer to first move it back into the database and then delete its entry in the dropbox group. In that way the dogs entry in the pets group works as it did before (the only difference is that the file to which it refers is in the database rather then on dropbox.


Thanks for the suggestions. I think I can solve these issues by just using duplicates. It is not often that I need to move stuff to dropbox. If I duplicated to my indexed dropbox, then send to external, I will have an “independent” file. Once deleted in Dropbox it should only be deleted in the indexed dropbox. Then the original file is still intact.

Thanks again for your suggestions

Yes, good idea ! :smiley: That will do (except when you use the dropbox folder to annotate your files :frowning: )

Thanks for the tip. I’ll remember it the next time when I help a colleague with DevonThink problems.

I index every folder and file in Dropbox and Box to one or more DEVONthink database. (For example, I index my Notational Velocity folder in every database.) I’ll echo some of @arnow’s advice, but not exactly:

  • replicate from and indexed group but never to an indexed group

  • if a document in an indexed group is deleted inside DEVONthink, then the group is refreshed in DEVONthink, the document will not reappear.

So, taking those two “rules” here is how to replicate the OP’s problem: index a folder “A”; replicate something in the folder to another group “B”; delete the instance in the first group “A”. The instance in “B” that was a replicant is not longer a replicant, it looks like it’s indexed, but it isn’t.

This situation drove me batty a few years ago until I realized what was going on. So, I formulated those two rules for myself, and a third:

  • if an indexed folder needs reorganization, either move everything into the database and reorganize there, or delete the indices, reorganize in the filesystem, and reindex.