Sync to folder just ,messed up my project.

I am new to the synced to fold er process and I might have done something wrong, but I don’t know what it is.

After a full day of working in the external folder on three major files, I opened my project, and let it sync. To my dismay, the entire project appeared in the updated files collection.although I had only touched three files and had made no changes whatever in the filenames. What was worse, the entire binder got shuffled: many files in the wrong order, and some files appearing as children of other files that had nothing to do with them. I have several hundred files in my project, arranged in logical folders, so this mixup is a major blow to me.

To avoid any further damage. I cleared the sync process and closed my project without syncing. Even if the files in the external folder remain in something like the correct order, the fact that it is a flat list means that it will be of only limited use reconstructing my binder.

I think I will now have to abandon the current project, go back to yesterday’s backup and start over, replacing manually the files I worked on today. But I obviously can’t to that until I find out what happened today to make sure that it doesn’t happen.

TIA for any suggestions.

This sounds a lot like this case. Are you trying to use this feature to sync two different projects together? If so the internal numbering will nearly always be different and so you end up with the wrong content being jammed into binder items.

It probably would be best to step back to your last backup, depending on how widespread the shuffling is—or maybe open the backup alongside the current version and use it as a reference to restore data (you can drag and drop between binders). Even if you do have to roll back, the content in the folder sync files should be up to date, and can be merged manually rather than automatically synced.

Hi Amber, thanks for your quick reply.
My situation is similar to the one you referred me to, except that I do not use dropbox or any cloud system. I sync between the project on my desktop hard drive and an external folder on an ST card that is sometimes used in, my laptop. All the work I did yesterday before the disastrous sync was done on the desktop.

I think I have discovered the immediate cause of yesterday’s disaster, but not the root cause of my problem. A few days ago. the program changed all the date modified entries on the files in my external folder to the current date of that sync. Therefore yesterday the entire project appeared in the updated documents list, because the dates on the folder were obviously later than what existed in the project.

So what I really need to know now, is why those dates were changed and how to avoid that in the future. And did that cause the shuffling of my binder?

I am going to have to rollback through several updates and backups and then tried to restore the work that I have done in the interim. I would like to continue to use this feature because my fading eyesight requires me to work w screen magnification program which doesn’t perform properly within Scrivener. Help, please.

Okay, the additional ingredient of Dropbox in the other thread might be misleading in the sense that this kind of problem between two projects can be produced using any means to transfer the files from one computer to another. In the case of Dropbox it is using the Internet and a bunch of automatic stuff, but it is essentially the same as copying the files yourself by hand using a drive.

The main issue is to not use this folder sync feature with more than one project. It is designed to be used so you can use other software entirely (most often something mobile, but there are many ways to use it).

That would make sense from the standpoint of it syncing more than it needs to, since modification dates are how it decides whether a file should be synced. However that alone would be fairly “safe” in that even if it syncs unnecessarily, if there is only one project using these files, the content would still be the same in the file.

I can’t say how to avoid the modification date problem—that depends on what did it.

I can say that wouldn’t cause any shuffling by itself. What will cause shuffling is if one project writes out a file as “10 - Name of item [40].rtf”, and that “40” part is what tells the software which part of the binder it comes from. If another project syncs from this same folder, it might use “40” internally to refer to a completely different file. But it can’t know that, all it sees is a file that should update 40, so the wrong content ends up in the binder.

If you have enough of that going on, it could indeed make a big mess. That shouldn’t happen with text editors that just open files individually, because they don’t care what it is called. The only way you’d get shuffling is if you opened several files and actually cut and paste stuff from one file into another and so on—deliberately you might say.

If you’re saying this is happening and you aren’t using two copies of Scrivener with one sync folder, then the problem is more mysterious. The only thing I can think of that might cause a problem like that is if one duplicates a project on the disk, and they both are using this same sync folder, and one then later on accidentally opens the old one and it syncs—but it’s using old internal numbers and so cannot properly use the newer sync folder. That’s a variation on the “two project problem”, and I’m not certain that would actually happen, but it’s a theory.

What I would recommend for a bit here is to use the File ▸ Back Up ▸ Back Up Now or Back Up To… command before syncing, and turn off the auto-sync on open-close, so that you can better control when it happens, and have the most recent version of the project to return to if it goes wrong. At least until the origin of the problem is discovered. Once things are smoothed out, it can be trusted to go automatic again.

Couldn’t you simply use the zoom feature in the Editor, and also increase the font size in the Binder in Settings, to overcome that problem?

Amber, you were right. I have only one Scrivener project that I’m working on, and I assumed I had only one version of it – that every time I clicked on the Scrivener logo to open it, I would get the same version. But your post prompted me to dig a little deeper and discovered, to my chagrin, that I had at least three operable versions of the project on my hard disk, and probably several more. (By “operable” I mean a “project name.scriv” folder that if clicked on will open the project directly. Some of these seem to have come from a backup that was opened for inspection and somehow remained open, out of the zip, and therefore in play. But the provenance of others is not clear.
At least one version contains the old correct modified dates, and another seems to have preserved the proper binder sequence. But I’m not sure which is which. So the mixup is even more tangled than I had realized. It will be difficult and time-consuming to repair this and I will probably need your further advice as I go along. Please bear with me.
First, I want to prevent any further sync. I believe the best way to do that is to rename the external folder. Correct? And to prevent any new backups until it’s all fixed, do I have to turn off automatic backups on each version that I open, or is there a global method?
Next, I have to browse through all the versions to find the one that I should repair, using some elements from the other ones. Is clicking on a “Project name.scriv” folder directly, instead of clicking on the Scrivener logo, the best way to look at one version at a time and to be able to record which is which? And what is the proper way to open a zipped backup folder (in order to inscpect it or to restore a lost file) and then to zip it up again so that it doesn’t leave an operable version behind?
In the sub folder where I thought my true project resided there is a subfoler called "Project name.scriv " which is type-leabed as a simple file folder. When I open that I see the usual scriv subfolders, (files, settings, Docs etc.) and also another “Project name.scriv” folder that is type-labeled “Scrivener Project.” Clicking on that one opens a version of the project. Do the normal visible Scriv subfolders represent the contents of the active, operable project in the same folder, or are there two different versions of the project right here? The answer to that will obviously affect how long it will take to straighten this out.
Many, many TIA

Thanks, lunk. I;ve tried those thinks. But main problem is that my mag program has a very sophisticated and helpful reader feature that doesn’t read Scrivener It can magnify any portion of the screen, but it doesnn’t know what it is looking at. Out of Scriv, it is very helpful.

What in Scrivener is it that can’t be adjusted so you have to use the magnifying program? Or phrased another way. Exactly what is it your magnifying software can do that Scrivener can’t?

Yes, that is definitely the first step I would take—I should have recommended that before. If a project (any project) cannot find the sync folder it will disable the feature.

In Tools ▸ Options… there is a “Backup” tab where you will find a global switch. Just remember to turn it back on once you have things sorted out! :slight_smile:

You can also individually exclude projects with the File ▸ Back Up ▸ menu, which you can do right after opening them.

Another thing I would do in Options is in the General section. I would switch off the feature that automatically loads projects when you start the software. Until you get things pruned down to one working project again, that mechanism may work against you.

And yes, directly loading the projects, once you switch that off, will be the only way to load projects.

There is no need to do that given how zip files work. You extract a copy of the project, open it to examine it, and if it is not good you can simply trash the project folder. The .zip doesn’t go anywhere, and rezipping the one you extracted will only clutter things up because now you have two .zips of the same project.

That is not a normal scenario if I understand you correctly. There should not be projects stored inside of other projects.

lunk, this is a known problem with Scrivener that was discussed here in the forum many months ago in threads regarding the dictation program Dragon Naturally Speaking, which also cannot “read” the words in a Scrivener file within Scrivener. It has to do with the manner in which Scrivener stores data on disk,which differs from the way other programs do it and therefore limits compatibility. This syhcing feature that I am trying to set up on my system was apparently designed as a way to work around that Scrivener attribute. I t seems there are many programs that face the same problem. My luck is that two of the programs I am using more frequently because of my fading vision, are on that list.

Amber, thanks for that very useful info and guidance, especially in making clear how I can open one project at a time.

I’m still repairing my sync mess. I had six versions o f my project, all of those created since I started syncing are deficient to some degree. But I found on my backup flash drive a zip backup dated the day before I started syncing, and I am going to roll back to that.
I can’t just delete the others because some of them contain newer data.
So how should I proceed to update the good one manually? Before unzipping the good one can I, should I change the names of all the others to avoid confusion? Or is it possible to keep two projects with the same name open on the disk at the same time?
And before I start: if I open the good zip where it is the project, I assume, will open on the flash drive. Is it best to copy the zip folder first to where I want the project to live, or move the project later?

TIA for all suggestions.

As long as each version of the project lives in a separate spot, you should be good to go – however, that can be confusing.

You may want to open a new, blank project in the destination you want it to be in and name it something different. Then, open up the version of your project you are going to save the most out of (we’ll call this version one) and copy and paste it all into your blank project. Now save the (new, non-blank project). Rename version one and move it out of the way.

Open up version two (the next version with the next most amount of stuff to save). Copy it over to the new project. Now rename version two and move it out of the way.

Lather, rinse, repeat, working your way through each copy in descending order of “I need to save this stuff back into the new project”.

Save your six old versions somewhere else (like an external drive, etc.) and remove all versions of them from your drive. At this point you should have the unified new project containing all of the relevant material you’ve recovered.

Thanks, Devin, for your suggestions, but I don’t think they would work for me.
Version 1 is clean and undamaged; it just needs updating. All the others are damaged in some way by the syncing disaster, a week or so ago. If I copy then one by one on to version 1, that would really spread the damage to my one good version. What I need is a safe way to keep two versions open at the same time so that I can compare the good one to the others one at a time and make the needed updates without perpetuating the errors. From what you say, though, I assume I can do that as long as the two projects are located in different folders, even if they have the same name. Is that correct?

If it were me, here’s what I would do:

One by one, rename each copy of the project to make it clear what it is. ‘MyProject28Sept.scriv’ is better than ‘MyProjectV1.scriv,’ and either of those is better than having six different folders named ‘MyProject.scriv.’

Create a folder called ‘ProjectRecoveryBackups.’

Open the project that you want to use as the starting point, which will probably be the one with the most ‘good’ component files. I’ll call this one GoldenProject. Make a backup to the recovery folder you created above.

With GoldenProject open, open one of your other copies. (I’ll call it WoundedProject1.) Figure out what files you want to keep. Drag those files into GoldenProject. Make another backup to the ProjectRecoveryBackups folder.

If you want, you can compare the individual files at this point. That is, look at both the original GoldenProject Chapter A and the copy you dragged in, and decide which one to keep. Or you can wait and do that later. If you do the comparison now, make another backup when you’re done. Close WoundedProject1.

Open WoundedProject2, and repeat, including the backup steps. Repeat as many times as necessary.

When you’re done, GoldenProject will be complete and up to date, and you’ll have made a separate backup at each step of the recovery process. Make one final backup, and get back to work. You can delete the intermediate backups now, but I’d recommend waiting until things have been working correctly for a week or so, just in case.


Thanks, Katherine, that’s the kind of solid support I was looking for. It seems like a very solid and safe procedure that even I will not be able to mess up. starting on it immediately.

Hi All,
I’ve v just been stymied again, this time by a new problem.
I can’t access the project version I want to use as my basic good one, what Katherine called the Golden project. It exists as a zip backup on a flash drive. The zip opens but when I get down to the SCRIVX file and click it I get a Location Error, telling me it cannot access the file I want, but with the file name preceded by a long path going back to C:/users’… Then it says: “The file is not writable: Access denied. Auto-saves need write permission to your project.”
I’ve tried it with Admin user, from different computers and drives, checked over permissions as best I can, but no luck. I have opened files from that flash drive’s back ups in the past, and I don’t know what is happening now or how to

TIA for all suggestions.

Did you copy it off of the flash drive before trying to open it?

If that doesn’t help, go here: … ng-project