I took your top-voted questions from https://pollgab.com/room/brento. At one point, my camera overheats – hang in there, I get it fixed in a few seconds.

Here’s what we discussed:

• 00:00 Start
• 01:59 MyTeaGotCold: Your HA/DR worksheet mentions a $1,000,000 option for 0/0 RTO/RPO. Where can I learn more about how this?
• 04:43 Trushit: It was good to see you at PASS. What is one things that you learned at PASS that made you go “that sounds cool!”? Anything interesting you want to share about PASS?
• 05:19 UserName Fabricated: Hi Brent, Have you done much performance work with the Fabric Data Warehouse? I’ve seen some articles that suggest you cannot create indexes and cannot see execution plans. Does this mean performance is so good that they are not required?
• 06:15 Jack: Hi Brent, I’ve changed the fill factor on a 40m row table from 90% to 100%. After rebuilding the indexes, the table size got bigger when I was expecting it to be smaller. Is there a reason for this? Thanks
• 07:18 Bandhu: We don’t have a dev SQL Server (Azure SQL VM) with customer data so the engineers code and test on production SQL server with live customer data. Customer data not allowed outside of production. What bad stuff have you seen from this scenario? Any recommendations?
• 08:14 Luis C.: Hi Brent. In what scenario is it recommended to deploy more than one instance of SQL Server in a production environment? How are resource conflicts such as CPU, I/O access, and network bottlenecks handled? Regards.
• 09:15 Elijah: There are hundreds to thousands of applicants for the positions I find on LinkedIn. How would you recommend getting a new role in this market? I have 5 years of MSSQL experience, 2 of those include Aurora work, and am looking for development (not production DBA) related roles.
• 11:23 Lewis Hamilton: Hi Brent, have you considered doing a podcast for office hours?
• 12:21 Jessica: When you were in Iceland on the Digital Nomad Visa, did you have any issues with language barriers?
• 13:49 rlidzydba: Hello Brent. I have an extended event to capture information about failed login attempts. I want to have an alert fire a job, but I am finding the data isn’t in the extended event file when the job runs. What is the expected delay after a failed login occurs?
• 14:47 DBA_Veil: Hey, Brent. Wondering about linked servers to created to themselves. SrvA would have linked server SrvA. If the local linked server does exist and someone queries it, does that process go out the network and back or is SQL smart enough to stay local? Asking for a developer…
• 16:03 Yoga_Blocks: Hey Brent, I’m a new hire (Dev role) at a medium sized company. We don’t anonymize customer data for our dev server. Does the push to do some come from or from security, or compliance? Is it my responsibility to bring this to management’s attention?
• 23:16 SportsFan101: Whats the current best practice for creating a dev environment or getting dev’s access to test data? Our Dev’s laptops have more resources than our current dev server.
• 25:14 I’m a potato ?: Hi Brent. I’m trying to reduce disk latency, specifically, tempdb files are 255 Avg Write Stall (ms). I’ve seen that you can move tempdb to a local SSD/ephemeral disk. Can you use it in a virtualize envelopment with Live migration/vMotion enabled?