Getting error in live solution hosted in azure recently. SAVE is not working in the backoffice. 20000+ errors are logged. One or more errors occurred. (Exceptions were thrown by listed actions. (invalid deletion count: 2 vs docCount=1 (resource: BufferedChecksumIndexInput(SimpleFSIndexInput(path="/home/site/wwwroot/umbraco/Data/TEMP/ExamineIndexes/ExternalIndex/segments_35"))))) Your help is much appreciated.

Hi there! I would recommend doing the following things: First go to the settings section and Rebuild your in memory content cache Then go to the examine manager in the settings section and rebuild your indexes. Then try again

Also make sure you use the recommended index storage settings for Azure Web Apps: https://docs.umbraco.com/umbraco-cms/v/10.latest-lts/fundamentals/setup/server-setup/azure-web-apps#recommended-configuration

I have verified the recommended appsettings. It has been added. I rebuilt the memory cache from backoffice and the examine management option in the backoffice is failing. This is a live solution running for months. I had same issue two weeks back. I deleted the folder from azure via KUDU and rebuilt the index. It worked. But I dont want to redo it . I wanted to know the root cause of the issue.

If the management dashboard doesn't load, then you need to delete the index files from the filesystem in your hosting environment

I deleted the examine index files. The index is still having issue. In the logs, a warning is logged in the ""No last synced Id found, this generally means this is a new server/install. A cold boot will be triggered."" Is this connected to the corrupted index error?

I don't think it is related

Yes. Restarting the application did fixed the issue. The issue occured second time this month. I would like to know the reason for the corrupted index.

I couldn't tell you. I have similar issues from time to time. It usually occurs when creating content in code or when doing bulk actions on content or when indexing takes longer than it should or when you loadbalance the backoffice. There's a multitude of possible reasons.

ok. Thanks for your support @D_Inventor @Jemayn

Best explanation of what it might be I have seen is from @Nik. The examine manager gets itself stuck in a unloading state and can't recover

I noticed that when I restart the app in Azure, the index get corrupted. Any idea why it is corrupting the index file