I guess the table scan is seqid from db2. Strange that is isn't an index scan. Is that the table or is that the one of the top-clusterindex scan

The not-in...

As Ken said, stall's are usually an infrastructure thing. (network, san, ...)

Could also be due a new query plan generating a heavier load

The server is cpu-bound -> find the most...

As Phil said, provide an easy digestable  testcase. So it can run on a test system to reproduce the solution. You've already shared a screenshot

You can mock the tablenames/values as...

Isn't a hardware fault.

Resource_semaphore:

SQL Server decided on an execution plan that requires more memory.

http://blog.sqlgrease.com/how-to-fix-resource_semaphore-waits/#:~:text=Typically%20RESOURCE_SEMAPHORE%20waits%20occur%20as%20a%20result%20of,need%20tuning%20%28i.e.%20adhoc%20queries%2C%20new%20code%20releases%29.

https://www.sqlskills.com/help/waits/resource_semaphore/

Add more server ram or  since you're on sql server 2017, you can use query...

I gues as many times your subreport is called ( the distinct values?) + 1 for the main report

Not sure if it's related but occasionally we have corrupt subscriptions. Those have a double datasource

check

select 
DataSource.DSID DatasourceID
,DataSource.Name DatasourceName
,Subscriptions.[SubscriptionID] SubscriptionID
,Subscriptions.Description SubscriptionName
,Reports.Name RerportName
,Reports.Path ReportPath
from [ReportServer].[dbo].[DataSource] DataSource
left join [ReportServer].dbo.Subscriptions...

Thanks for the start article. Installed the libraries from the resources in the SSRS GAC. Also copied them to ReportServer\bin folder.

Todo: reportbuilder

Mixed feelings about the logging. Sometimes it's just "Unexpected termination".

Wow, running sql server 2005. Maybe it's better to setup a new 2019 and restoring the data. TLS 1.0 en 1.1 are getting obsolete fast.

Regarding linked server: did it work...

MS knows the answer and MS can vouch for it. Best route is to contact MS

As far I can read is that if you have SA (subscription) you're allowed one...

Lately there was some visibility because of an Apple OS bug. We've found a rather annoying bug that allows remote takeover, patch asap

And the implemented solution is / was?

That's why we are still using VM's. Containers / K8 are next level in management

I'm not familiar with AAG groups or replica's.

As you're working with AAG's you might get funding for a MS support call / consultant to sort it out

I would also go on the route of permissions. Perhaps you're not seeing all the files. (difference between user and "run as administrator")