Thanks guys for taking your time to read this and your replies!

Got my doubts cleared out.

Have any of your SQL Instances run out of threads and "crashed" temporarily but then...

Thanks for the quick reply Jeff!

Do you use Auto update stats? And for critical tables full scan with norecompute? And update these weekly? Or just full scan all stats with...

Thanks guys for taking your time and reading the post, I'm gonna look up all the options and make some tests, also i'm going to pay more attention to the...

Thanks for your reply Jeffrey, so you would basically create a shared storage on either server of the cluster and use this one as a witness? and if it can...

Yes, that may turn out as a deadlock and SQL Server will kill one of the sessions so the other can finish.

https://blog.quest.com/the-anatomy-of-sql-server-deadlocks-and-the-best-ways-to-avoid-them/

No, it should not affect the job unless the job is touching any database in the primary replica which you want to reboot but thats not the case here.

But i...

• This reply was modified 2 years, 6 months ago by  Alejandro Santana.

Have never used transactional replication to be honest.

but heres a document from Microsoft, take a look at it.

SQL Server Agent jobs may fail after you change the...

use the catalog views.

in the view sys.server_event_session_fields the name of the values are filename and metadatafile, match them with sys.server_event_sessions i think the id.

Regards,

The plan is okay.

If you want to shrink the log, you gotta use the dbcc shrinkfile.

I don't think you should be shrinking the log file if its still increasing in...

That's actually a good idea Jeff, thanks 🙂

In our case when it happens its usually in old applications and we want to know which user is using it since in...

The way we do it is that we don't check when was the last transaction, we put an extended event capturing anything that goes into the database and we let...

Check where the sessions is connecting from, if its from SQL Server itself, then you have to wait like the guys said earlier.

Sometimes its an outside connection like service connecting...

Yeah i completely understand what you mean.

the instance was working fine, transaction times started taking longer in night and what "fixed" the issue at last was increasing a data file...

The cost threshold is actually 25 but the queries/transaction running in the instance never go above it so they are not actually using parallelism, i've checked most queries and none...

Where i work we also have a backup tool, we have access to the backup tool dashboards and we have access to the policies (execute,modify,delete and create) of backups of...