Looks like "runs fast in ssms but slow in application"

check ms documentation https://learn.microsoft.com/en-us/troubleshoot/sql/database-engine/performance/troubleshoot-application-slow-ssms-fast

With modern storage I would skip index maintenance as Jef mentioned in that thread. So many layers and can be quite costly in the cloud

Most important is having correct statistics....

Sounds like what I've read earlier today about automatic tuning

Also, when a plan is forced , it is allow to compile longer?

https://littlekendra.com/2024/01/17/automatic-plan-forcing-could-be-great-but-isnt-sql-server/

https://www.brentozar.com/archive/2024/08/video-i-must-be-an-idiot-automatic-tuning-never-works-for-me/

https://dba.stackexchange.com/questions/339748/azure-sql-managed-instance-1-excessive-initial-compile-times-leading-to-app-ti

Dateformats ?

Empty or Invalid "To" when emailing ?

The latest thing thrown around is MS Fabric.

Found a page about the differences between Fabric and Dataverse https://fabriconelake.com/microsoft-fabric-vs-dataverse-comparison-for-microservices-and-data-management/ 

Our company is not that far yet to represent information as an unified...

Luckily not, didn't know that CrowdStrike existed before it happened

search on "token-based-server-access-validation-failed-with-an-infrastructure-error-login-lacks-connect-sql-permission"

most articles mention logins & service principal names.https://blog.sqlauthority.com/2015/08/19/sql-server-login-failed-for-user-reason-token-based-server-access-validation-failed-with-an-infrastructure-error/ Mentions checking for deny permissions

That's how we did it. Most important things are the encryption keys

I've found https://www.sqlservercentral.com/articles/migrating-sql-server-reporting-services

To migrate from 2014 to 2016. 2016 to 2019 should be similar.

Don't have experience with scale-out.

Beware, from 2017/2019 MS disabled multiple SSRS instances on the same server

As we work on goodwill, please provide some sample data, the formatted query (like Phil did), expected results

DDL: tablestructure/views

Sampledata: insert statements

Query: formatted / human readable ( code tag?)

Expected results

Otherwise: have...

LOL "NOTE: No Refresh Buttons were abused in posting this message"

Nvarchar(max). Every now and then the developers get a reminder to review those. They sneak in when you're in a hurry.

SELECT table_schema,table_name,COLUMN_NAME
FROM INFORMATION_SCHEMA.COLUMNS
WHERE data_type in ('nvarchar','varchar','char','nchar')
and CHARACTER_MAXIMUM_LENGTH=-1
ORDER...

Thanks

Steve for the Json references, Phil for the validation, Frederico for the date suggestion.

My workaround is storing it in an nvarchar(max) variable, replace the wanted character and write it back...

Thanks Steve, that explains it

We're currently using to much on-premises addons (sql agent, SSRS, SSIS, linked servers). Currently spoiled with our hardware infrastructure.

If we move, it would be Azure SQL for small databases.