I'm experiencing an issue with editing data through Portal web maps. When I make a lot of edits to multiple points in the same layer in a short period of time (few minutes), the layer I am editing quits loading (all of the other layers load, however). The layer either does not draw at all, or only a few points draw, but not the entire layer. I'm watching the internet traffic during this and it completes the request, there are no time outs or anything like that.
When going to the server logs, I receive the following message: "Error: The requested operation is invalid on a closed state [layer name][STATE_ID = ######]" (The pound signs represent an actual number, but that number is dynamic).
I've seen this issue on multiple feature services that contain data from multiple EGDB's. The only way to resolve this issue is to restart the feature service. That immediately fixes the issue, but re-appears again after making a few edits.
I'm running ArcGIS Enterprise 10.5.1 and Microsoft SQL Server. The layers I am editing are registered as versioned.
Any ideas on what could be causing this and how to prevent this from happening?
I reconcile and compress every day due to the amount of editing we are doing to our EGDB's. In the SDE_compress_log, the end_state_count doesn't always get down to 1. But when troubleshooting this issue, I did disconnect all users, reconcile and compress, and get the end_state_count down to 1, and the issue still exists.
I have also the same issue especially when editing the polygon in web map that failed to perform the editing operation giving the same error in server logs. Did you fix yet?. Please let me know if there is a better way to tackle this if you have fixed this issue?
I contacted ESRI support and I was told that is was because of lack of resources (RAM) on the server that contains the EGDB. We added more RAM, but it did not resolve the issue. I did not contact ESRI after adding additional RAM. The only way we were able to resolve it is move the EGDB to a different server. It's possible that we still don't have enough resources or it could be a configuration issue with the server. Regardless, moving the EGDB to a different server solved the issue.
After moving a production database to a brand new VM and having web maps pull from it, I am now receiving this message again. While web map performance has significantly increased, we still receive this error message (although the frequency of this message has decreased). Regardless, this error message is severely hampering our ability to deploy web maps to users as this causes a large productivity drop because the web maps are practically unusable until I either restart the feature service, or the user waits about 15-20 minutes. With this happening multiple times per day to multiple people, the lost time adds up quickly.
This message appears to be happening on layers that are edited a lot on the web map.
Does anyone have any solutions to this?