Are category lists attached to each data file or the individual Quicken ID?

ozzy032

I'm using Quicken 2017 H&B. I have two separate Quicken data files. One to track regular monthly finances and accounts (File A) and the second a to track retirement accounts (File . I recently deleted most categories not investment related from the category list in File B. Now, when I do a One Step Update on File A, the category list is replaced with the one from File B. All existing transactions that were given a category like 'Groceries' or 'Auto:Fuel' are now shown as Uncategorized because the category was deleted during the update. How can I resolve this? Now, If I make any changes to the category list from either data file, it now migrates to the other. They are basically syncing for some reason. I have two separate data files for a reason. I'd like to keep their categories lists separate too.

splasher

Each Quicken data file is self-contained.  If categories are missing in both files, then you deleted them yourself.

Proof of what I am saying is that you can import and export the category list from a given file (File menu) and if the categories were global, there would be no need for that process.

ozzy032

Category list is syncing between files. To verify, I restored File A from backup and categories were there. I immediately ran a one step update, and then they were all gone. I then added a dummy category to File A. Then, when opening File B and running the one step update, the dummy category added in File A shows up in File B.

Some additional information is that these data files used to be a single data file. I recently copied the file and the deleted some of the accounts from A and the other accounts from B.

splasher

Are you using QMobile/QWeb?

ozzy032

Not that I'm aware of, no.

Tom Young

What you're describing simply should not be happening, period.  Each separate file has its own unique Category list and closing a file, opening a different file, and then performing a OSU shouldn't change the Category list in the newly opened file. The fact that the two files are using the same Quicken ID doesn't affect that. 

I'd suggest contacting Quicken support directly

https://www.quicken.com/support#windows

and seeing if you can talk this through on a phone call.

splasher

I would restore the data file again and then use File->Save a copy as to break the connection of the two files which Q somehow thinks are the same.
I have never heard of what you are describing and I have made lots of Windows Explorer copies of files for testing.  Make sure when you open the copy, that you do not activate Q-Mobile.

ozzy032

Tom Young, I agree. It is extremely bizarre. I'm scratching my head as to why this is happening. I'm literally making a change in one file, updating, then closing file and opening the other file, updating, and then viola! there's the change. Super frustrating. I'll give splasher's suggesting of restoring again and then saving as something else to see if that a does anything.

splasher

One way to tell if QMobile has anything to do with this is to check:
Edit->Preferences->Mobile&Web
Does it show that it is not setup or is not set to sync?

Chris_QPW

ozzy032 said:

Category list is syncing between files. To verify, I restored File A from backup and categories were there. I immediately ran a one step update, and then they were all gone. I then added a dummy category to File A. Then, when opening File B and running the one step update, the dummy category added in File A shows up in File B.

Some additional information is that these data files used to be a single data file. I recently copied the file and the deleted some of the accounts from A and the other accounts from B.

Somehow you are syncing to the cloud.

I did the following test.

I took my data file that isn't syncing to the cloud and made two copies using Windows copy and paste.

Then if the first copy I deleted a category that I use a lot, and so the transactions no longer have a category.  Then I did One Step Update.

Next I opened the second copy, and then did One Step Update, and it didn't change, the category is still there and filled in, in all the transactions.

Next I made two copies of a data file that I have for testing Quicken Web/Mobile/cloud sync using Windows copy and paste.

I repeat above with the exception that I just used "Cloud Sync" instead of One Step Update, because this data file doesn't have any accounts setup for downloading.

The result was that on the sync to the second copy the category was gone, and missing from all the transactions (as expected).

Note that Data file to  Data file isn't "supported" because of all the ramifications of what may go wrong, but in reality they can't "block it".  Each data file has an unique Id embedded in it to tell them apart.  When you use Windows copy and paste or File -> Save a copy as... you have an identical data file, and Quicken can't tell the difference between what you did in one from the other.

Note the main reason that Data file to Data file isn't supported is because the copy on the server (cloud account) is a partial copy.  So that means that it is possible to change something in data file A sync and the cloud account won't get it, and as such it can't sync that data to data file B.

ozzy032

Any suggestion on how I can turn off cloud syncing? I don't want to use that function anyway. I'd prefer all data only be saved on my hard drive.

Sherlock

ozzy032 said:

Any suggestion on how I can turn off cloud syncing? I don't want to use that function anyway. I'd prefer all data only be saved on my hard drive.

If you haven't already, you may want to review: https://www.quicken.com/support/what-quicken-cloud

splasher

@Chris_QPW
I thought  File->Save a copy as wasn't an exact copy with a different file id whereas the File->File Operations->Copy kept the same file id, but did the copy at the transaction level.

ozzy032

An update after discussing with Quicken support. Quicken assigns a unique 18 digit ID to every new data file created. It's called a dataset ID. Since I had created one of these files by copying a previous file, these dataset IDs were the same. For some reason, despite having sync to cloud turned off for all of my files, Quicken still uploads some data to the cloud (like category lists I guess). The only way the tech was able to help me was to direct me to create a completely new blank data file, and then export all data from the old file to the new using QXF transfer. This mostly worked, except for with my two brokerage accounts. I'm told that it will work for those two accounts too after I upgrade to 2020.

Thanks for the comments and suggestions.

Chris_QPW

splasher said:

@Chris_QPW
I thought  File->Save a copy as wasn't an exact copy with a different file id whereas the File->File Operations->Copy kept the same file id, but did the copy at the transaction level.

@splasher I'm pretty sure it is just the opposite,  File->Save a copy as gives the identical file and File->File Operations->Copy is the one that changes the file unique Id.

But now that I think of it I got this wrong.  the "data file Id" is used for Express Web Connect.  It isn't used for Quicken web/mobile/cloud account.

For the cloud account it has a "dataset name/Id" and neither of these copies changes this.

File Operation Copy

Save As Copy:

And note that on the preferences dialog there is an option to rename it.  I tried it and without even syncing it changed the name in the other file, which is what I expected.  There is a "hidden dataset Id" that is really what they use to connected the data file to the cloud account/data set.  They have to do this so that even if the user uses the same name in a different data file it won't conflict and sync to the wrong data file, which would be a disaster.

Given this I don't think there is any way to change the "cloud account data set Id".

The only thought I have left on this is to try to turn on the cloud sync in both data files, and then turn it off.  BTW one thing you could check to see if they are "linked", is just what I did, as in rename the dataset name in one and see if it changes in the other.

Edit -> Preferences -> Quicken ID & Cloud Accounts -> Edit dataset name.

splasher

Guess what, with 27.1.26.14, you can't delete the cloud data file any longer.  
You can turn off Sync though via Preferences.

  

     

splasher

I was just advised that I might be wrong on my last past if syncing is turned on.
Since I don't/won't turn on syncing, I can not verify.

Rocket J Squirrel

Chris_QPW said:

splasher said:

@Chris_QPW
I thought  File->Save a copy as wasn't an exact copy with a different file id whereas the File->File Operations->Copy kept the same file id, but did the copy at the transaction level.

@splasher I'm pretty sure it is just the opposite,  File->Save a copy as gives the identical file and File->File Operations->Copy is the one that changes the file unique Id.

My understanding is that there is no way to change the Dataset ID. That ID is created only by File > New and remains in all copies no matter how the copies are made.

A few versions ago, you could see the Dataset ID by doing

Edit > Preferences > Cloud Accounts > CTRL+Edit cloud account name

but that command disappeared when the preferences recently changed. If you have an older version of Quicken lying around, try it.

[EDIT: the command is still there, but does not appear until the QDF file is connected to QCS.]

Chris_QPW

OK fill in a few questions statements from above.

(Note I keep a different data file around with basically nothing in it to do this testing of cloud operations.  I don't do it with my real data.)

Yes, both "Reset your cloud data", it really makes no sense to have these if you are syncing.

Reset cloud (delete and rebuild):

Delete cloud data from another file (you can't do the current one):

Note every data file has a cloud account because it is used for different things other than just web/mobile data, like the credit card score.  If you delete a cloud account from another data file and then open it, Quicken will recreate the cloud account when either you do a One Step Update or log out and back in.

On "Ids"  was purposely trying to give them different names even though I think I have heard Quicken Inc use the same term for two different things.

Express Web Connect predates the Quicken cloud account by many years.  It has a "file Id/data set Id" that is used for basically the same purpose keeping track of what data file is syncing with those accounts.  And this is the Id that I believe can get changed by the File Operation -> Copy, but in truth it almost never matters because it seems that the Express Web Connect system has a way to "invalidate" the data and recreate and as such has really had any problems that really needed it changed.  The subject only came up when it was found that there was some "cross talk" of updating the same Express Web Connect account in two different data files.  As if one updated in one, the second won't download those transactions.  I haven't seen this problem in years, and it would be something completely different.  I made the mistake of connecting the two.

Now on the Dataset Id that is used for the cloud account.  I used RJS' "hidden key sequence" on the latest Quicken and it works:

So now we have way to see it, but no way to change it.

Chris_QPW

P.S. the two "Dataset Ids" might be the same, but that isn't what  Quicken Tamara said years ago.  At any case it doesn't seem to matter.  Express Web Connect seems to have gotten around whatever problems it had in this area, and clearly none of the copies change the this Dataset Id that is used for the Quicken cloud account connection.

ozzy032

For what it's worth, I can confirm that Quicken Support used the same 'hidden key' sequence to view the Dataset ID for the file. They also stated that it was not possible to change this number, and any copy operation creates a copy of the file with the same Dataset ID. My only option was to create a new Quicken data file using the File>New Quicken File which generated a data file with a new unique Dataset ID. I then had to do a QXF data export/import.

Chris_QPW

@ozzy032

Yes that makes sense.

BTW looking back and your comment, I see I missed something:

"The only way the tech was able to help me was to direct me to create a completely new blank data file, and then export all data from the old file to the new using QXF transfer. This mostly worked, except for with my two brokerage accounts. I'm told that it will work for those two accounts too after I upgrade to 2020."

This is incorrect.  Quicken 2020/Subscription QXF import doesn't support investment accounts either.

I say import because it is known that you can export a QXF file and import it into Quicken Mac and for the most part the investment transactions transfer.  So clearly the investment transactions are put in the QXF file.  But they have never put that support into the Quicken Windows QXF import.

splasher

@ozzy032 @Chris_QPW

Got the following response from Quicken Sarah:

@splasher - If a brand new file is created, it is assigned a new 18 digit dataset ID automatically.
Exporting/importing data via QXF only moves the data, not the associated dataset ID which remains tied to the original file.
If a User wanted a new/different dataset ID assigned to the new file, this is accomplished by deleting the cloud account associated with the new file.  Once deleted, a new cloud account with a new 18 digit dataset ID is automatically created.

So, there is a way to change the dataset ID without re-creating the file.

Rocket J Squirrel

splasher said:

@ozzy032 @Chris_QPW

Got the following response from Quicken Sarah:

@splasher - If a brand new file is created, it is assigned a new 18 digit dataset ID automatically.
Exporting/importing data via QXF only moves the data, not the associated dataset ID which remains tied to the original file.
If a User wanted a new/different dataset ID assigned to the new file, this is accomplished by deleting the cloud account associated with the new file.  Once deleted, a new cloud account with a new 18 digit dataset ID is automatically created.

So, there is a way to change the dataset ID without re-creating the file.

Verified. I had to open a different QDF file, then delete the real QDF file from the cloud account, then reconnect the QDF file to my existing Quicken ID. After all that, the file had a new Dataset ID.

Furthermore, I discovered a new way to show those IDs. In the Quicken ID & Cloud Accounts preferences, hold CTRL while clicking Cloud accounts associated with this Quicken ID.

Sherlock

@Rocket J Squirrel  

If you restore a backup of the datafile saved before the dataset was removed, does the restored datafile also get a new Dataset ID?  

Chris_QPW

Sherlock said:

@Rocket J Squirrel  

If you restore a backup of the datafile saved before the dataset was removed, does the restored datafile also get a new Dataset ID?  

Without even checking that seems impossible to me.  It implies believing that the data set Id isn't in the data file.  The whole point of the data set Id is to uniquely identify the data file, therefore the Id has to be in the data file.  And so if you restore an old data file, it has to have the old data set Id.

Sherlock

Chris_QPW said:

Sherlock said:

@Rocket J Squirrel  

If you restore a backup of the datafile saved before the dataset was removed, does the restored datafile also get a new Dataset ID?  

Without even checking that seems impossible to me.  It implies believing that the data set Id isn't in the data file.  The whole point of the data set Id is to uniquely identify the data file, therefore the Id has to be in the data file.  And so if you restore an old data file, it has to have the old data set Id.

@Rocket J Squirrel appears to have confirmed the Dataset IDs are maintained in the Quicken Cloud associated with the Quicken ID as, when we delete a Dataset ID in the Quicken Cloud and open a datafile with the original Dataset ID, a new Dataset ID is created and associated with the datafile we just opened.  What I'm asking is, if we open a different datafile which still has the original Dataset ID, does this datafile also get a different new Dataset ID or will all of the datafiles with the original Dataset ID be mapped once again to a shared Dataset ID?

I used the example of a datafile backup saved before the original Dataset ID was deleted to expose the potential flaw in the design.  One could continue to restore the datafile backup creating many unreferenced Dataset IDs in the Quicken Cloud.

Separately, I noted the Synclog appears to be indicate deleted Dataset IDs persist.

Rocket J Squirrel

Sherlock said:

@Rocket J Squirrel  

If you restore a backup of the datafile saved before the dataset was removed, does the restored datafile also get a new Dataset ID?  

I don't know. I don't use the Quicken cloud and I am testing this on my test machine, where I generally do not perform backups.

I did find one backup on this machine, though. When I restored it and connected it to the cloud, it got a new Dataset ID. Surprisingly, the restored backup file appears to have replaced my original QDF in the cloud. Compare the first row in this image with the first row in my image in the post above.

Sherlock

@Rocket J Squirrel  Thanks.

So the good news is, we do have a way of splitting the files linked to same Dataset ID.  The bad news is, we split all the files and backups linked to the original Dataset ID.  I can foresee some interesting issues...

Chris_QPW

I don't think you can exactly say the "data set Id is stored in the cloud account" it isn't really the correct way to look at it.

What is needed is a "connection" between the data file and the cloud account.

One common way this kind of thing happens a username.  When you want to log into Quicken.com you give it your username.  That is how the server knows to pull up your account data.

If you think of the data set Id is your user name, then you see that Quicken needs to present its "username/data set Id" to the Quicken cloud server so that it can get the right account data.

So the data set Id has to be stored in the Quicken data file.  And as such a copy of that data file is going to have the same data set Id.

The "cloud account" certainly has the "data set id", but more than call it "stored in" it is more the "key/index" on the server to open the cloud account data.

Sherlock

After performing my own tests, I suspect there is a distinct file identifier maintained in the file across all backups and copies, etc. being used to identify the original file.  I still know of no way to separate the files.

There appears to be many ways of producing new Dataset IDs but there is still only one entry for the file being maintained in the Cloud.