Select to view content in your preferred language

Edits will not synchronize

2308
12
10-05-2013 10:38 PM
TeresaSmithson2
Emerging Contributor
Replication: One way parent-child with archiving
Parent Database: ArcGIS 10.1 SDE geodatabase on SQL Express on a dedicated server
Child Database: File geodatabase on network server

Feature class structure (simplified):
   Feature class of buildings - contains attribute for building number and polygons
   Feature class for building data - featureless (no geometry) attribute table containing all of the information regarding the building.
   Feature classes for other information - featureless attribute tables for plumbing, electrical, etc.
The "featureless" feature classes are not "tables", they are feature classes.  I found the search ability to be more robust this way, but I also had synchronizing problems when this data was just tables.

Relationships: all of the various tables are linked via the building number or other common attributes between tables (e.g. building data has a code for plumbing that links to the plumbing table.  Many of the links become circular, which works well for complete coverage of the data.

Problem:
If I edit any of the cells in the buildings feature class, the changes will synchronize.  If I edit any of the cells in the featureless feature classes (data only attribute tables), none of the changes will synchronize.  Is there something in the way that the tables or database is setup that would cause this problem.  It's a good thing I have short hair - harder to pull out:-)

Any suggestion is wlcome, I have tried even things that seemed silly; all to no avail, but I am something of a newbie to this stuff.
Teresa
0 Kudos
12 Replies
847396730
Frequent Contributor
Good morning! It may be your first rodeo, but you have discovered the truth about GIS: it stands for "Professional Troubleshooting!" 

I have something for you to try: replicas hate invalid geometry.  The introduction of invalid geometry will fail a replica.  I suspect this is what you're experiencing.  To find out...

Open an mxd, and add the parent data. Start an editing session.  From your editing toolbar, click the Editor drop-down, and click Validate Features.  It will likely pick up all the features in your "featureless feature classes," and tell you they're invalid.  This would explain the failure; an invalid geometry introduced after the replica is created will fail the replication process.

The way I get around it: fix the invalid geometry, or recreate the replica and allow the invalid geometry to persist.  In your case, the actual design is creating the invalid geometry, so you would need to either rebuild every time you synch, or switch over to geodatabase tables, and hit the forums for a solution to the search/table fail issues you described.

I hope that helps!
0 Kudos
TeresaSmithson2
Emerging Contributor
Good morning! It may be your first rodeo, but you have discovered the truth about GIS: it stands for "Professional Troubleshooting!" 

I have something for you to try: replicas hate invalid geometry.  The introduction of invalid geometry will fail a replica.  I suspect this is what you're experiencing.  To find out...

Open an mxd, and add the parent data. Start an editing session.  From your editing toolbar, click the Editor drop-down, and click Validate Features.  It will likely pick up all the features in your "featureless feature classes," and tell you they're invalid.  This would explain the failure; an invalid geometry introduced after the replica is created will fail the replication process.

The way I get around it: fix the invalid geometry, or recreate the replica and allow the invalid geometry to persist.  In your case, the actual design is creating the invalid geometry, so you would need to either rebuild every time you synch, or switch over to geodatabase tables, and hit the forums for a solution to the search/table fail issues you described.

I hope that helps!


Hi Marianne,
I have eight tables that are data only; all but one returned valid.  I am going to look into why that one table would be different from the other 7 since they were all done the same way at the same time.  But I don't think this is the result you expected.

Appreciate your help.
0 Kudos
847396730
Frequent Contributor
That does surprise me.  I'm sorry that didn't help.  The last step I would try is setting up some small testing replicas, and figuring out exactly what does and doesn't work.  I will be interested to learn the solution, when you find it!
0 Kudos