0

I have an enterprise geodatabase with many feature datasets. I was editing one of the feature datasets and accidentally deleted some records from there. I did not realize and stop editing after saving it. When I opened my attribute table, I do not see those selected records.

Is there anyway that I can retrieve those deleted ones in the appropriate data sets?

  • 2
    We need a number of additional details: which version of ArcGIS, which RDBMS, whether the edited feature class were versioned, whether any reconciliation or compression has occurred. You should also sequester the active set of backups, so that you can recover the data if necessary. – Vince Apr 3 '18 at 14:51
  • It's ArcGIS 10.4 and SQL server. I do not have idea about versioned and reconciliation or compression. – BPS Apr 3 '18 at 15:47
5

You have a few possible options and they all depend on your specific environment and set-up. First, if you have versioning turned on for the feature classes in question and you have not reconciled and posted all of the versions, you could connect to (or have another user connect to) the database with a different version that the edits were not made in and that has not been reconciled and posted since the bad edits, and they should still see the features. You could then export those features, connect back to your version, and append those features back in. Additionally, if the only edits you made in your edit version were the mistaken edits and you have not reconciled and posted them yet, you could always just get your database administrator to delete your edit version from the database and re-create it, effectively removing all of the mistaken edits. And there are some other options with versioning such as having another user edit the same features and using reconcile and post settings very carefully (by row not attribute for example) and reviewing conflicts and resolving as desired to retain the features. More on versioning http://desktop.arcgis.com/en/arcmap/10.3/manage-data/geodatabases/an-overview-of-versioning.htm

If versioning solution is not an option, perhaps you have archiving turned on for the datasets in question. If so, it is as simple as connecting to the database using a historic version time prior to when the bad edits were made. Then you can select those features, export them, connect back to the live dataset, and append in those features. More on Archiving http://desktop.arcgis.com/en/arcmap/latest/manage-data/geodatabases/geodatabase-archiving.htm

Next, if it is an enterprise database, that means you are using some sort of database management system, such as Microsoft SQL Server, Oracle, PostgreSQL, etc. And most of those systems have various back-up procedures and options. If you had your system set up to run some database backup, you should be able to restore a backup of the database, either over-writing the current database and returning it to a backup before the bad edits were made, or restore the backup to a new database you can connect to and extract the features in question out of and then append them back into the live database.

Finally, depending on your situation, it may be worth looking through (and/or having other users look through) some of their analysis work they may have done with those datasets. For another, unrelated, prior project you or someone else may have run a clip geoprocessing tool (just for example) against the dataset(s) in question, covering some or all of the features in question, and the output of that tool may be in some project specific file geodatabase on your system somewhere. If so, you should be able to go into that dataset and again, get the features and copy/paste or append them into the main live datasets.

Again, these procedures will vary depending on your local set-up. So use with care (and if none of them work, look into setting up some of the mentioned options so you can handle similar situations in the future).

  • Thank you, John, I am going to check every possibilities. – BPS Apr 3 '18 at 15:43
  • @BPS If your organization has a Database Administrator, now would be the time to reach out to them if you're unsure about any of these database settings, backup procedures, etc... If you're not sure who that would be, then maybe start by asking whoever it is that you would ask to grant/deny you view &/or edit permissions on a dataset in the database, or who would set up new users on the database. They may be a good person to check with first if you have such a person available. – John Apr 3 '18 at 16:44
  • You are right. I am going to explore these things. Its a nice stat to know more about the data-sets. Thanks. – BPS Apr 3 '18 at 16:52

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.