The Backup Wrap-Up

Julie Ulrich, Systems Engineer at Farm Bureau Insurance of Michigan, joins us on this episode to talk about her experiences with NetBackup and Rubrik in her world. She’s been working in backups for over 25 years, so has seen a number of iterations of both products. We talk about many of the challenges she had with NetBackup that led her to considering Rubrik, as well as the pros and cons of using Rubrik. We also talk a little about her concerns about Microsoft 365.

These are the kinds of stories that keep you up at night. UK police deleted hundreds of thousands of records. Sysadmin accidentally deleting thousands of users. A new backup "feature" that made the backups worthless. The infamous story of Toy Story getting deleted with no backups! All this and more!

In another fascinating look into a very different world (backup-wise), we are delighted to have Timo Piiparinen from Multicom in Finland. He’s been in the IT industry for over 40 years at the same company! TImo gave us a fascinating look into a very different backup world. He makes a case for tape in the backup system (something I gave up on a while ago) because he’s using a backup software vendor that didn’t give up on the medium and actually innovated for it. Using a unique multiplexing setup that used flash disk as a big cache for the tape, very large block sizes, and what he called read-optimized writes, they run these tape drives at over 650 MB/s during backups and can guarantee restore speeds! He and I spar a bit over the value of disk during DR, and his position was that this was only necessary for the most critical systems. He’s a fan of tape, which is hard to find these days. TImo will be back.

We welcome Andrew Wertkin, Chief Strategy Officer of BlueCat, a DNS, DHCP, & IPAM (DDI) security company. Like backup, properly securing these parts of your infrastructure are both extremely important – and everyone thinks it's boring. I knew nothing about DDI before this recording and I learned a ton about the ways that bad actors use these technologies to either attack or exploit your company. Using technologies like Bluecat can actually stop ransomware in its tracks! Andrew explains exactly how ransomware attackers use DNS for Command and Control, and how products like Bluecat can be used to stop them in ther tracks. This is a great episode with a lot of really good information.

Vanessa Toves joins us again to explain exactly why you need to backup Microsoft 365 and similar services. We talk about how Microsoft is only responsible for that platform; you are responsible for the data. Particular attention is given to the idea that somehow Retention Policies are a substitute for backup. She explains exactly why that is not the case. In fact, her experience with such policies has her struggling to wrap her head around why someone would want to do that.

Vanessa Toves, a Microsoft 365 expert, joins us on Restore it All to explain the unique architecture of Microsoft 365 (formerly known as Office365). We talk about Teams and Groups, and how many people misuse both. We cover Sharepoint's role in this as well, and how Exchange Online figures into the picture. This was originally going to just be an overview, but our architecture questions just kept coming, and Vanessa kept answering them. This episode lays a good foundation of what we will cover in the next episode, which will be dedicate to backing up Microsoft 365.

Prasanna and Curtis wax philisophical at the end of 2020, and try to predict what they think will happen next year. We look forward to the end of 2020 (don't we all?) and look with a positive outlook to 2021. We talk about the vaccine, and what it's going to be like getting back to work, and whether or not there will be a new normal there that's not like what it was. We talk about SaaS products and their prevalence, as well as the emergence of a number of backup products based on Kubernetes. 2020 has been quite the year; let's hope 2021 is better.

Here is the FTC page I referenced in one of the editor's notes: https://www.ftc.gov/tips-advice/competition-guidance/guide-antitrust-laws/single-firm-conduct/predatory-or-below-cost

Fathom Analytics (https://www.usefathom.com) is a privacy-first analytics engine, and they were attacked in November 2020 via a very big DDOS attack that threatened their core business. Jack Ellis, co-founder of Fathom Analytics, is a developer and first used the developer mindset to address this threat. He quickly realized he was out of his depth and decided to call for help. The AWS Shield Advanced team came to the rescue and helped them shut down the attack, and helped to stop additional attacks that were still on the horizon.

Jack wrote about his experiences in the blog post "Someone attacked our company," and its URL is below. Now he shares his experience with us. What's it like to be attacke by an unknown assailant bent on doing your company harm – and to not be sure how to stop it? Join us on this episode to find out!

Someone Attacked Our Company: https://usefathom.com/blog/ddos-attack

We get a closer look at the eye of the storm. A few episodes ago, we had an anonymous guest we called Harry Potter, who told us what it was like to manage a DR remotely. This week we hear from the man on the ground, recovering two datacenters after a hurricane wiped out the island where they resided. We learn all sorts of lessons, especially about how so much of a disaster recovery has very little to do with what backup people think of when they think about a "recovery." In fact, we spend almost this entire episode NOT talking about backup! We talk about network connections, places to sleep, how to eat, and how to recover your datacenter when the neighborhood power won't even stay on. This guest is also remaining anonymous, and since he is Harry Potter's friend, we called him Ron Weasley. (My apologies to the Potterheads out there.)

Cornelia Davis, 30-year industry veteran and CTO of Weaveworks, explains what cloud-native design is and why it matters. She explains that you do not have to go fully cloud-native on day one if you are moving an app to the cloud, but everything you do "refactor" to a cloud-native architecture will be more resilient, scalable, and cost you less to operate. We talk about Kubernetes and how that figures into everything, but how it is not the only way to be cloud-native.

Listeners to the podcast can get a 40% discount on Cornelia's book by using our code "podrestore20"

https://www.manning.com/books/cloud-native-patterns?query=Cloud%20Native%20Patterns