How to back up Jira

Start listening


Jira is yet another service that could be at the center of your organization, and losing the data stored in there could cost you a lot. Did you know it doesn’t even have an audit trail for many things? Not only can you lose data, you might not know what was deleted or who did it! We are joined this week from two representatives from Revyz, a new service to back up Jira. They talk about what to back up and what you can do for free, along with what functionality that misses out on. They then explain how their new service works and what it offers. Great episode where I learned a lot. Even if you don’t use Jira, you will find a lot of useful info in this episode.

Transcript

[00:00:34] W. Curtis Preston: Hi, and welcome to Backup Central’s Restore it All podcast. I’m your host w Curtis Preston, AKA Mr. Backup, and I have with me, my cylinder misfire detection consultant, Prasanna Malaiyandi, how’s it going Prasanna?

[00:00:48] Prasanna Malaiyandi: I’m good Curtis. Honestly, I think you should just throw the entire thing out and just start over.

[00:00:54] W. Curtis Preston: I don’t think that I don’t think that’s economically

[00:00:57] Prasanna Malaiyandi: No that don’t do that, but this is the ongoing saga for our listeners of Curtis’s car that has an on again, off again, cylinder misfire issue, which may or may not be a head gasket issue, which may or may not be other things.

[00:01:11] W. Curtis Preston: Yeah, I either have a blown head gasket or, or a head gasket that is blowing That is in the process of being blown, uh, or an EGR valve stuck in the open position. And. Uh, the usual and the thing is it’s, it’s an intermittent problem. And so troubleshooting is very, very difficult, you know, for those of you that are, you know, your, your computer geeks, right?

Like how do you fix a bug that only happens once every thousand times that you use the software?

[00:01:45] Prasanna Malaiyandi: exactly. That’s when you that’s, when you basically turn it into cos.

[00:01:50] W. Curtis Preston: to what?

[00:01:51] Prasanna Malaiyandi: Cosmic. I, one of the

[00:01:52] W. Curtis Preston: Oh yeah,

[00:01:54] Prasanna Malaiyandi: right. If a bug was hit and no one’s able to reproduce it, figure it out, right. It’d basically be like the stars aligned. There was a solar flare, some cosmic event happened and therefore the bug is no longer there.

[00:02:08] W. Curtis Preston: I’m down to, I’m going to do the repair of one or the other, and both of them are, are significant. And then see if it goes away. Right.

[00:02:18] Prasanna Malaiyandi: I thought the EGR is less work than

[00:02:21] W. Curtis Preston: No, it’s a lot. It’s actually quite a bit of work. Um, because you,

[00:02:24] Prasanna Malaiyandi: out to get to it.

[00:02:25] W. Curtis Preston: yeah. It’s. Yeah. And then, and then, and as long as you’re pulling the EGR valve, you should pull out the, um, the intake manifold and clean it out.

There’s some ports that need to be, I’ve been watching, I’ve been watching a lot of YouTube videos Prasanna. You’d be very proud of me.

[00:02:40] Prasanna Malaiyandi: I am now. What? But wait though, Curtis, are you watching at two X?

[00:02:44] W. Curtis Preston: Uh, no, I I’m old school, man. I watch it at one X. I actually like you. like that. Well, I’m not like you, I’m not watching like 19 hours of it.

[00:02:55] Prasanna Malaiyandi: Hey,

[00:02:56] W. Curtis Preston: to get,

[00:02:57] Prasanna Malaiyandi: like four hours. Come on.

[00:02:59] W. Curtis Preston: uh, yeah, but you’re watching 19 hours videos in four hours

[00:03:03] Prasanna Malaiyandi: Ah, I wish they would actually go faster than two X that’s. Now my feature request for YouTube. Any developers out there at YouTube?

[00:03:11] W. Curtis Preston: I can’t imagine what in the world you would do at two X. Anyway. So, um, well we’re gonna get onto something much more exciting than car maintenance. We’re gonna talk about backup today. You know, my favorite topic, we have two guests today. Our first guest started his career, just like me as a backup admin, I first cross pass with, with him when he was leading the security and compliance team at Druva, uh, he is a co-founder of Revyz.

Welcome to the podcast Sanket Parlikar

[00:03:41] Sanket Parlikar: glad to Glad be on the podcast.

[00:03:45] W. Curtis Preston: And we also have your co-founder. He started his career designing chips at SGI and has worked for multiple security firms, including Symantec, Druva, and Agari data. He’s also a co-founder at Revyz. Welcome to the podcast Vish Reddy?

[00:04:02] Vish Reddy: Thank you. Uh, very nice to be here.

[00:04:06] W. Curtis Preston: sort, sort of be here. What, what time is it over there right now

[00:04:11] Sanket Parlikar: It’s like 9 45, 9 50

[00:04:14] W. Curtis Preston: in the evening?

[00:04:15] Sanket Parlikar: Yep.

[00:04:16] W. Curtis Preston: Okay, well, thanks very much for, for joining us to talk about backups at basically 10 o’clock at night. I know I do that a lot, but usually I’m just talking to myself at

[00:04:26] Prasanna Malaiyandi: Socks like this is just me getting started for the evening.

[00:04:29] W. Curtis Preston: I remember it’s interesting. I remember 10 o’clock for me, like historically in my career, 10 o’clock was when I was doing backups, like for people, like 10 o’clock for some reason was the time that a lot of them seemed to kick off.

So this was like the beginning of the night for, you know, for a lot of, for a lot of time. But, uh, so anyway, so thanks for both of you coming on here. We’re we’re here to talk about. A new company, which you two, uh co-founded, which is called. I make sure I, I have pronounced it correctly. I’m assuming it is Revyz.

Did I, did I get that correct?

[00:05:07] Sanket Parlikar: Yes, that’s right.

[00:05:08] W. Curtis Preston: Okay.

[00:05:09] Prasanna Malaiyandi: But it’s an interesting spelling. So if you go to look

[00:05:11] W. Curtis Preston: Yeah, it is. Yeah. It is an interesting spelling, which is, which is, which is why I wasn’t sure if I was pronouncing it correctly. It’s R E V Y z.io as the website. Um, so why don’t we sort of start at the, and, and by the way, I, you know, I’ve already brought up Druva, uh, that’s where I happen to work.

Uh, I’ll throw out our standard disclaimer. Our standard disclaimer, Prasanna works at zoom. I work at Druva. This is not a podcast of either company. The opinions that you hear are ours and, um, be sure to rate us at your favorite podcaster, just give us stars, give us, and we really love comments. And currently there’s an offering on the play.

You guys didn’t even know this, but apparently if we. Nine new comments on apple, on iTunes. Um, you know, in the next week or two, then apparently I have to grow a

[00:06:04] Prasanna Malaiyandi: was, it was a month. We said

[00:06:06] W. Curtis Preston: Oh, that, oh, was it a month? It was a month from the last episode. Right?

[00:06:11] Prasanna Malaiyandi: went

[00:06:12] W. Curtis Preston: Right. So, uh, if we get nine new comments on apple iTunes, then I have to grow a beard for Christmas. Which, you know, I mean, it will still not look as full as Prasannas cuz it’s currently a, theard, is it a theard now?

[00:06:30] Prasanna Malaiyandi: uh, not quite a theard. It’s still a tweard. It’s like a and a.

[00:06:37] W. Curtis Preston: A two and a half year. That’s not, that doesn’t make any sense. Yeah. Uh, but yeah, cuz I, I, I, you know, I’ve really never grown a, a real beard and anyway, so there you go. So rate, you know, gives some comments and um, also if you wanna join in the conversation, we’d love to have you. So, if you like the topics that we talk about, backup security, privacy, uh, you know, ransomware, all of those things.

And, uh, then please just, uh, reach out to me at Twitter @wcpreston or w Curtis Preston at Gmail. So let’s go back to the beginning. What made you wanna do this crazy thing of, of founding a company? Well, I’ll, I’ll start with Vish here. Um, you know what, you know, what, what problem was out there that you said?

I, I think I can solve this problem.

[00:07:33] Vish Reddy: Uh, Curtis having met multiple customers at Druva having worked there. One of the things

[00:07:44] W. Curtis Preston: Mm-hmm

[00:07:47] Vish Reddy: applications, companies were deployed were reciting the cloud. You know, the cloud vendor, again, no fault of the cloud vendor, uh, you know, the cloud vendor would always say, Hey, Mr. Customer, it is your responsibility to secure your data, which is, which makes sense. Right? I mean, if a user and delete their data accidentally, you, you can’t hold the cloud vendor to be responsible for that.

[00:08:18] W. Curtis Preston: Well, can I interject on you there? Um, I mean, you’re telling your backstory, but I, I just have to interject there. I would love it. If the cloud vendors would say that. Right. I would love it. If the cloud vendors would make it very clear whose responsibility it is to protect the data UN unfortunately many of them don’t right.

They, they, they either, they either just don’t talk about it. Um, You, you know that that’s probably, that’s probably the worst to me is it is if they say nothing, it it’s not in their service contract, it’s not in their SLAs. And, and yet they may have legions of fans and, and, and I’m gonna, I’m gonna call out Microsoft as being the worst offender here, because they have legions of fans who say, you do not need to back this up.

They have not publicly clarified. They have a few things hit, you know, hidden here and there, but there is no. You know, um, public statement. So, you know, on the, on the reverse of that, I would give, I would put Salesforce. They, they, you know, not only have they clarified what your responsibilities are, they now offer their own service to back up.

Um, and then somewhere in the middle is, is, you know, 4,000 other SaaS vendors out there. Right. Um, but yeah, so. Uh, let’s come back to that. I wanna, I wanna come back to that. I want you to finish sort of your story, but I wanna come back to this topic. Uh, anyway, sorry. I was just, it’s a hobby horse. I just had to jump on.

So.

[00:09:51] Vish Reddy: Yeah. So, you know, that led us to saying, okay, this is a problem. Lot of, to your point. Lot of customers don’t know that, uh, there is, uh, in fact, uh, I’ll come back to a story. One of the first. Uh, you know, prospects or validators who we met had the exact same point as you, right? You said, why do we need to back?

You know, the vendor is providing the service. I’ll get into the details in a bit of that, but that was the Genesis of saying, okay, you know, there is a problem. When we looked into some of the more popular, um, DevOps focus, we were more focused on DevOps. You obviously, you mentioned office 365 Salesforce.

There’s a good ecosystem of vendors there. So we said the things people are looking at, which critical for the functioning of that business, uh, is important. Um, Workday servicenow. And there’s a growing list of, uh, applications, which are out there. And we said, okay, can we build an architecture? Which is, which can, you know, help protect that data?

Can we have a generic architecture, which can, where we can plug in any of these different applications and we can help simplify the job of the administrator. So that’s where we got started. Uh, that was the Genesis when we recognized this, that there’s this problem. Uh, we said, okay, you know, again, backup as you guys are more familiar, this is one of the basic functions, which it administrator performs now.

Unfortunately,

Now, there are some people who spent hours together just started to do this. And he said, that’s not, that’s not in today’s day and age. You should not be spending a lot of time doing this. This should be just offloaded. Should be at the end of the day. I dunno whether you guys agree with me or not. Backup is like insurance, you need it some point of time, but you never know when you need health insurance.

[00:11:56] W. Curtis Preston: Yeah. It’s, it’s exactly like insurance in that. You must buy it before you need it. you, you can’t get car insurance after you’ve had an accident. It’s exactly like insurance in that same way.

[00:12:10] Vish Reddy: Yeah. So just to continue the story. Oh, um, uh, you know, Sankit and I used to work together. Um, and, uh, we would exchange every so often we would exchange some ideas and say, oh, maybe we should do this. We should do that. And then I shared with him this idea, oh, why don’t we? Well, you know, there is a problem out here.

I’ve talked to a couple people and. know, they’re outlined that there’s don’t solve.

[00:12:48] Sanket Parlikar: Oh, yeah, it was a, a really. Journey, right. Uh, since beginning like, uh, inception itself, uh, exchanging ideas, talking to different people, trying to understand what is a real problem, uh, out there. Right? Uh, luckily, uh, I had a lot of, uh, admin friends, uh, who were managing different applications, uh, and then, uh, wish, uh, had different connections and all the people.

Talk to, uh, basically they said, wow, that’s amazing idea. Right? Uh, I do have a lot of critical data sitting into, uh, these, uh, DevOps tools. And specifically, if you talk about Atlassian, it’s again, uh, pretty sensitive data for me. Why don’t you help me, uh, back that up today? I have a problem, but I cannot predict that.

Uh, there are a lot of challenges, so exchanging a lot of ideas, talking to a lot of people. Uh, that’s how the journey, uh, started and then slowly validating that idea, validating that problem out there and then, uh, finding a solution for it. That’s how it all started.

[00:13:53] Prasanna Malaiyandi: Yeah, it’s great that you actually have like real world customers who have this problem. Right. A lot of times. You’re like, oh, I have this idea. Let me try to find a good market fit for this. Right. And it’s like, you’re struggling. But actually having a problem that people are struggling with is like, gets you so much further ahead.

The one thing, uh, and I don’t know, Sanket or Vish who wants to answer this is maybe Sanket is, I know you mentioned Atlassian. A lot of our listeners may not know what Atlassian is or who they are. Could you maybe give sort of a brief background about some of the products, um, that they might offer that people might be more familiar?

[00:14:30] Sanket Parlikar: Absolutely. So in general, uh, let’s talk about software industry, right? Uh, in software industry, typically, uh, developers need a system to track their work, right? Uh, they are developing the software. They would be logging in, uh, certain tickets to track, uh, track their tasks and work, uh, and how they, uh, develop a software.

Right. So typically, uh, what, uh, This software is, uh, is going to do for developers is, is going to create tickets. It’s going to manage all their work, their work log, and then build reports out of it. That’s how, uh, the engineering function, uh, tracks their efforts right now, Atlassian as a company, what they do is they focus on, uh, uh, creating a software.

Does that job for engineering function. And one of their application is called S JIRA software, uh, which will, uh, give that functionality to the engineering, uh, group, uh, wherein they can create different tickets, uh, log in their daily tasks and track it as, as they make the progress. So that’s the JIRA software.

There’s another application, for example, which is called as a confluence wherein you can go ahead, create your documentation, be your internal documentation. Be it your customer facing documentation. You create all the documentation and then use it for different purposes. So that’s what Atlassian does at pretty high level.

[00:15:54] Prasanna Malaiyandi: So it’s basically building all the tools you need rather than developers having to spend time focused on how do I track what’s going on in spreadsheets or in like documents, writing out things, right. It’s all easily available. So people have business. Gotcha. Across the

[00:16:09] W. Curtis Preston: So this was a, this is a tool that you were obviously familiar with in your various, you know, basically doing your regular job.

[00:16:18] Sanket Parlikar: Exactly. And, and I just gave you one example of software industry, right? Uh, because all of us come from software background or a product background, but when we connected with different, uh, people, different customers, uh, we discovered that it’s not only limited to software industry. Ticketing system in general is applicable to almost every industry.

We went to healthcare people, they’re using ticketing system. We went to automobile industry people. They’re using ticketing system in one way or the other. Right. So this is not limited to one specific software industry, but it’s applicable everywhere. Right. And yeah, people may use different tools, but the problem is still, uh, pretty generic.

[00:17:02] W. Curtis Preston: Right, right. A and so you talked about Vish, you talked about, um, wanting to develop a generic architecture. And I’d say that’s probably the biggest challenge that you, that you have is that you, when you’re developing a backup tool, that’s gonna back up so many different types of things. Um, you know, how do, how do you develop a, an architecture that you think will serve multiple tools like that?

[00:17:32] Vish Reddy: Right. I think that’s the, you’re absolutely right. That’s the, the biggest challenge we have and I think we’ve made significant progress in addressing the challenge. I’m not saying that, you know, we’ve completely solved that problem. Uh, at this point of time, uh, you know, we started off with Atlassian Atlassian, as Kate mentioned, is got JIRA confluence, JIRA service management.

It’s got a bunch of tools in itself. Uh, we’re sort of focusing on that, uh, uh, you know, challenge, so to speak. Uh, and then, you know, we’re hoping that the architecture have built out is gonna help us scale to others. JIRA while you know, ticketing system, it seems simple. You have a ticket, you have a user who’s submitting a ticket.

How, how complicated could that be? Right. That’s what would all think? But then once you look inside, if you look the, the database scheme for that, oh my God. That the first time I saw this, it was, it was a spaghetti bowl of like, you know, multiple tables interconnected with each other.

[00:18:35] W. Curtis Preston: I for the record, I would describe that as the JIRA user experience as well, but that’s but that’s, that’s just me.

[00:18:44] Vish Reddy: Right. So I think, you know, uh, the proof in, uh, using the product and seeing how far we’ll come, uh, come along now, one thing which I wanna also, uh, highlight here. So we, couple of prospects, we validated that idea. We actually went and pitched this to at. They said, Hey, you know, we not seeing anybody building something like this in your ecosystem.

We think that this is critical. Uh, is this what you guys and you know that between sun and

ICS would, we had had in 30 minutes, uh, the people who we’re talking to. Recognize that this is a good problem to solve for them. And they were like, yeah, you know, if you guys are willing to put in your efforts and solve this problem for us, we’ll, uh, you know, back you or we’ll fund you some amount of seed money to get started.

And not only that, they connected us back to their engineering team, the same team, which is developed that complicated, which is there. And, uh, we’re working very closely with them in, you know, uh, building more APIs, which will help. Um, you know, backup and restore data in a, in a much cleaner way. Uh, they do have good APIs, but those APIs are not built, uh, keeping backup and restored in mind.

Um, but they are working towards getting those new APIs out, uh, soon.

[00:20:15] Prasanna Malaiyandi: Yeah. And I think that’s one of the keys is a lot of the vendors, right. They try to stay agnostic and they’re like, Hey, here are the APIs. You go try to build your backup app or whatever app on top of it. I think the fact that you guys are actually getting engagement from the engineering team to be like, Hey.

This is really what it takes and probably also imparting that wisdom on, Hey, these are the requirements from a backup or a restore perspective, cuz that’s also the other challenges. Usually in like SaaS services, your restores aren’t very expansive, right? The options you provide, it’s normally like, yeah, I’m just gonna restore everything.

But very rarely do you have the ability to say restore a smaller object? Unless they’re really focused on thinking about that. Because typically as an engineer, you build features, you build product, you don’t always think about backup and restore data protection when you’re building these things.

[00:21:06] W. Curtis Preston: Well, I do, but, um,

[00:21:09] Prasanna Malaiyandi: and I know vision Sune do, but I’m just saying most people don’t

[00:21:14] W. Curtis Preston: no, I know what you, yeah. Yeah. I, I, and, and by the way, I, you know, I’ll, I’ll give credit to Atlassian because again, I’m gonna contrast to my favorite punching bag, uh, Microsoft. Right? They don’t, they, they didn’t have APIs for backup, right? The only way that you got, um, that we were, you know, that Druva and other companies were able to back up is basically use outlook web access.

Right. Um, and, and you use an API that was completely meant for meant for something completely different. And you’re using it for backup. I think now. For various reasons. My understanding is that Microsoft has come out with some, some new APIs, uh, but,

[00:21:57] Prasanna Malaiyandi: Not for everything though.

[00:21:59] W. Curtis Preston: Yeah. But not for everything. But yeah.

So, so, so hats off to Atlassian for giving you, you know, first, first off, acknowledging that this is a problem that needs to be solved. And, uh, and then giving you access to the APIs and to the, and to the engineers. That’s great.

[00:22:16] Prasanna Malaiyandi: I was going to jump back, actually, a couple of questions. Um, I know sanket, you talked about Atlassian, what they do. Um, I’m sure a lot of folks are asking though, why is a JIRA system or a confluence system? Like, why do I even need to back that up? Right. It’s just, oh, I created some things. These are all closed tickets.

Right? Why do I even need to back up this old data?

[00:22:40] Sanket Parlikar: Good question. So I will just go back to my previous example, software industry right now, engineering is working on a life project, for example, right. Uh, typically what they do is, uh, they try to put in, uh, all the, uh, comments and all the, uh, data, what they’re working on. In in, into that ticket. Right. And a lot of the times, even the, uh, version control system, uh, which is like bit bucket GitHub, it’s linked with your, uh, JIRA.

So now all the commits are being tracked in a ticket. Now, if that data is lost, the tickets, uh, present in JIRA, if those are lost, for some reason you lose that visibility of what happened. Yes. Uh, you have to, uh, go back, dig in, uh, put in more efforts to understand what happened, right? But your productivity is impacted. Your ongoing activity of, uh, whatever, uh, ongoing sprint or whatever it is it’s impacted.

It’s a direct business impact. So that’s one of the reasons why you need to protect that data and make sure you can roll back as soon as possible. Right? Another impact is basically the reporting structure. Lot of big teams, they rely on JIRA to, uh, understand what is going on within their scrum teams. So that’s very critical.

Right now that’s one example. The other example is lot of, uh, customers, they leverage, uh, JIRA software. They leverage JIRA service management tool for their HR function, uh, for different, uh, uh, functions like HR finance. They also leverage Atlassian apps wherein they store sensitive information. For example, what we have seen is HR teams.

They onboard, uh, different, uh, employees, uh, in, in, uh, Jira itself, right. Or JSM, they, uh, conduct interviews and they store certain, uh, amount of data employee data into, uh, the JIRA software and JSM tools. Right. So that is even more, uh, sensitive data. And, um, Very impactful for your business, if it it’s lost. Uh, so there are multiple angles to it.

Not only, uh, the, uh, business continuity part of it, but also how sensitive data, uh, uh, is, is being stored into these systems. And what if that data is lost? Right? So there are different, uh, angles to it.

[00:25:14] W. Curtis Preston: so let me ask this question. So Atlassian obviously has a big business already. Um, and they have a lot of customers. Certainly some customers have said, gee, I, I, I agree with everything that you just said. What would an Atlassian customer do today? If they wanted to get any kind of backup of that?

I’ll, I’ll give that to, to.

[00:25:38] Vish Reddy: Yeah today. Um, so, uh, let back up a little bit, which is Atlassian offers their products in two deployment, uh, modes, um, on premises. And cloud, uh, of course they came out their on premises product, you know, they. most customers when they, uh, who are very, uh, you know, sensitive about their data backups and so on for on premises systems, they would back up the database.

Uh, typically it’s a, MySQL, or a Postgress database. I think you have choices. So on premises systems, you would back up their database itself, but still the problem remains, I think, as you were pointing out, right. You typically go mess up with one or two things. You, you know, you don’t typically blow up your entire site most of the time.

Uh, now the problem with that, you know, backing up your databases, you do have point time snapshots of your database. So if you don’t roll that back then you do all the work, which you’ve done from the point of time, you know, you roll back to right. OK. So that that’s the current state of affairs on the, on premises side of things. On the cloud what Atlassian provided was a mechanism to do a database export. Uh, so in their UI today, most customers try to go and press pressing that button. Unfortunately, that takes a very long time. Again, depends on your size of your data, uh, your, your tenant, right? How much of data which you have, uh, if you have a couple of, um, you know, let’s, let’s go with gigabytes.

Let’s say you have 10 gigabytes of data. Uh, tickets, attachments, whatnot. That’s gonna take you some time to, uh, back up. Now, here’s the thing, which is really messed up in my, uh, there’s a cloud product. You gotta take that backup. You gotta download that your on premises system, and then you decide what to do with it.

You can put it back into the cloud and, you know, manage different versions. Um, or you can put it on a, on premises system and manage that. Right. But here again, if you wanna go,

[00:27:58] W. Curtis Preston: Is there a way to restore it to the cloud.

[00:28:02] Vish Reddy: yes, you can restore it to the cloud again. It’s all or nothing. So if you wanna go back, let’s say a month, that means you’ve lost a month’s of data. It’s not, you know, , it’s not like I want this one ticket or two tickets or this attach. Um, and it’s a lot of manual work by the way. Um, in the current state of affairs, if your data size is greater than five gigs compressed, uh, then you gotta go and, uh, you know, open that file up, split it up into multiple parts and then try uploading.

Uh, so there is a lot of work which you would have to do,

[00:28:43] W. Curtis Preston: Hmm,

[00:28:45] Vish Reddy: and this is.

[00:28:46] Prasanna Malaiyandi: It reminds me of like Salesforce backup and restores. I know

[00:28:49] W. Curtis Preston: It does. Yeah.

[00:28:50] Prasanna Malaiyandi: on right talking about that and just trying to deal with that seems like a bit of a nightmare vish

[00:28:57] Vish Reddy: Actually, I would say, you know, uh, one of our first, uh, prospects. They’re in the travel industry, uh, very different, they’re not software. Uh, they would, uh, log, uh, customer issues, changes to the travel schedule and so on and in JIRA. And, uh, they were doing this manually and they were looking for a solution like, every two.

Hello. And so they’re like, oh, there’s too much manual work. It takes at least five to six hours to just download the data and then you gotta manage it. I don’t wanna do this. Can I just offload this to. Um, I think that was the inspiration. I would say that was one of the, you know, stories. I was gonna tell you as to how we said, okay, maybe there’s a good problem to go.

[00:29:56] W. Curtis Preston: And so the, the, the worry, like, again, going to your, your travel example, the, the worry would be that you accidentally delete like just a customer. You know, you, you just deleted all history for a customer that you’ve been managing travel for five years, and then you just deleted that customer. Your only choice as you were saying is to restore the entire database.

You can’t restore just that customer, right?

[00:30:24] Vish Reddy: That’s correct. Uh, there is no, you know, pick and choose you it’s all or nothing. Um, one other thing, which is, uh, you important for, uh, to, and lot of people don’t this, uh, which is when you delete something in, in JIRA or confluence. It is gone. You cannot get it back. There is no audit trail for that.

[00:30:52] W. Curtis Preston: Really none.

[00:30:53] Vish Reddy: I I’ll give you another example.

[00:30:55] W. Curtis Preston: a recycled bin or anything.

[00:30:58] Vish Reddy: okay. So there are, there is a recycled bin. If you delete things at a project level, but if you delete things at the ticket level, there is no recycle bin. There is no audit log for it. Last year, I was working for, again, this was another data point, which we had, I was working for Agari data, uh, email security.

We got acquired by a company called help systems. And there were a lot of upset employees at Agari who did not like, you know, what was going on. And. I, we would use JIRA for our, uh, software development, project management and so on. I started noticing tickets disappearing, and I’m like, I’m pretty sure I saw this.

I can see AGA 129. I saw it yesterday, but it’s not there in the system. Where did it go? I go asking around and nobody knows because there is no trace of it anymore.

[00:32:00] W. Curtis Preston: Wow. That’s not good.

[00:32:04] Vish Reddy: And that is the state of affairs today.

[00:32:07] W. Curtis Preston: Right. Wow.

[00:32:10] Sanket Parlikar: So if, if I may add right, uh, Curtis, uh, how, how, uh, Backup and restore really works. Uh, 10 years ago or 20 years ago when you started, right? Uh, the focus used to be on disaster recovery. Uh, 30 years, about 20 years ago, it was all about disaster recovery. Meaning my system is up and running OnPrem system.

Something goes, uh, bad. I just rolled back, uh, probably, uh, five days older, a snapshot things work. Right. And I’m okay to give up some of the data now, uh, Uh, with cloud apps, basically the situation is like, Hey, I’m not ready to give up all of my data. Uh, I just want to, uh, pick, choose what I want to restore.

Right. Uh, because, uh, yeah, there are a number of reasons. I, I, uh, accidentally deleted something, right. Uh, why do I roll back, uh, entire tenant? I, uh, I had a malicious incident where in someone. Deleted only subset of data. Why do I roll back completely? Right. So these are the evolving scenarios. Now, malicious actors being the biggest one.

As we talk about security, that’s the biggest threat evolving for any SaaS app.

[00:33:25] W. Curtis Preston: Yeah, I, I, we actually had a discussion on the previous, uh, podcast that, you know, I, I just sort of had this realization that, you know, when I started, uh, which for the record was 30 years ago.

when I

[00:33:43] Prasanna Malaiyandi: was being kind to you.

[00:33:45] W. Curtis Preston: But what I started, the, the, I will agree that the primary thing that we were trying to solve was hardware failure.

Right. Whether it was, it failed because of a disaster or failed because we were running servers on a single hard drive, right? No raid, no, nothing, just a hard drive. And, um, I mean, if there were multiple hard drives, we were just using all of them as individual drives, but, and, and, and between the, the, the, um, the change in the technology with everybody using, uh, raid and things like that. And the change in SaaS providers, Dr. Is not my problem, right. With Atlassian with, with JIRA, Dr. Is not my problem. That’s their problem that we all agree that it is their problem to get the service up and running. That is what I’m paying them for.

But. Due to the, to the change in, in the way technology has, has evolved the number one reason. In fact, I, I would argue that like it’s like 99.9%. The number one reason for a restore is now humans, not hardware. And Dr. Uh, I mean, it’s still, there are still disasters, but the disasters are caused by humans.

They are they’re caused by. Um, you know, ransomware attacks and other malicious attacks, like what Vish was talking about. So I would, I don’t know if this was the point you were making, but I’ll make it if you weren’t making it and I’ll agree if you were making it. And that is that. You have, if you’re designing a backup product for this world, you have to design it with that, um, in mind, right?

Since the primary reason that we’re going to be doing restores is dumb stuff. right. We need to make the restore of dumb stuff. Uh, the, the easiest thing that the product can do. Did did I say the same thing you said, but in a whole lot more words.

[00:35:50] Sanket Parlikar: Exactly.

[00:35:51] W. Curtis Preston: Okay.

[00:35:52] Prasanna Malaiyandi: Curtis does that a lot

[00:35:54] W. Curtis Preston: I’ve literally made a career of doing just that. Um, so, so the, the companies knew, in fact, I, my understanding is that, you know, you’re, you’re, you’re just now launching what, are you handling all of Atlassian products?

I’ll give that to Vish.

[00:36:12] Vish Reddy: we got started in April, uh, for five months in roughly speaking, um, version one or MVP is gonna be covering JIRA Software. Um, and then we have a robust roadmap, uh, I think next would be Jira SM, Confluence.

SM

[00:36:38] Prasanna Malaiyandi: For your MVP or version one? Could you talk a little bit? I’m definitely sure you’re doing the backups. What about from the restores? I know earlier you talked about today with what you get with, uh, JIRA today or Atlassian today, it’s sort of, you get everything as a backup once every two days at most, right?

And then you have to restore everything back. What sort of restores do you handle?

[00:37:01] Vish Reddy: Great question. No, you know, what’s the point of doing backup without being able to restore back right?

[00:37:06] Prasanna Malaiyandi: Curtis and I were just talking about that yesterday.

[00:37:10] W. Curtis Preston: No one cares. If you can back up Vish

[00:37:13] Vish Reddy: So.

Which, and choosing that one thing or two things, or, you know, again, being very granular about what you can restore by. So that is part of our MVP. We back up every day, every 24 hours, automatic backup, remember this is insurance, right? You just buy it and you forget about it till you need it. Of course. Um, and so when you need it, you can go back in, you can go pick the specific day from which you want that backup, uh, or that, you know, piece of data, maybe it’s a ticket, maybe it’s an attachment.

Um, and then you can go and filter out, uh, and say, okay, I want this specific thing to be restored back. That’s what we do. There are lot more use cases, which are there as an example, something which we’re not doing in the MVP, lot of customers have multiple different sites or tenants. Uh, you know, you control five different tenants.

You may move things around. It’s sort of a data migration use case. So to speak, you’re backed up something in from one place. You wanna put it into another place. Um, that’s something that you’ll be doing in the future. Uh, uh, that’s not part of MVP today. What we do is whatever you back up from a given tenant, you can put it back into the same. Now JIRA to the earlier point, which we were making.

It is a pretty complex. The data structure is pretty complex. So there is actual data which consists of, uh, you know, uh, comments, potentially description of what the ticket is. And so on. And attachments, attachments could be drawings, could be code snippets, you know, various or zoom recordings. Lot of people add in zoom recordings have seen into the ticket itself.

Uh, that’s part of your data. Then there is a whole bunch of configuration configuration here could be workflows, could be, you know, um, you know, different screens would show up, uh, you know, to different users. Um, and so we back up both and we can restore both those things back. Uh, uh, we, for the MVP, we are, we are able to restore back everything on the data side of this.

Uh, anything you do with the issues, you can restore that back, uh, on the configuration side, uh, we’ve started off with screens and workflows because those, those to be the most important things based on ours. And then, uh, you know, over the course of the next two, three weeks, uh, we’ll be adding in more configuration elements to, uh, to be.

[00:40:00] Prasanna Malaiyandi: I’m glad you covered the config elements because that’s actually, one of the things I was gonna ask is, especially with SaaS services, people sometimes forget about like the settings and configs, right. That’s also critical to capture in terms of backup and restore because someone makes a change.

Hey, I wanna be able to restore that. Or like you said, Vish workflows, people don’t necessarily think of that as like a JIRA ticket, but it’s still important for the business.

[00:40:23] Vish Reddy: One of the learnings which we, uh, had was. Uh, you know, a lot of companies go through SOC2 compliance audit right now. So these companies also have JIRA as their change management system, which tells you, Hey, we, this, this point, this is what’s are just about to embark on getting ourselves SOC2 compliant. And in talking to the auditor, they were like, I didn’t know that, uh, you know, you, if you delete some tickets, they can, there is no sign of it anymore. That’s a big problem for

[00:41:13] W. Curtis Preston: Yeah,

[00:41:14] Vish Reddy: this is interesting, this, you know, as we’re.

[00:41:16] W. Curtis Preston: honestly, that that’s a, that’s a major hole for JIRA, but that’s a problem for them to solve. Like they, they need to have that audit log. Right? Uh, the, the, the story that you mentioned earlier, um, it, it should, you should not be able to just go in and delete tickets without record. Right. Um, we should be able to go, oh, it was Steve.

Steve is the one who deleted all the tickets that he didn’t want you to know about. Right. Um, but so sort of one final question, cuz we’re, we’re getting short on time, uh, earlier, I think it was Vish that mentioned that currently the, the, the built-in product, you can only back up every other day.

How is it that you’re able to back up every day? Is it because of the, the partnership that you.

[00:42:01] Vish Reddy: Great question. So. Uh, if you use the backup functionality, which Atlasian gives you, which is your entire database backup, then you can do it only every 48 hours. Uh, if you’re backing up attachments and so on, what we’re doing is different. We are using that APIs. We are going, and each of their tickets, uh, we’re being very granular, right?

And we’re not getting the entire data set in one shot. We’re going and picking everything one at a. And backing it up. Now, here is where things are. Um, and I think most people are familiar with this is, your first backup is gonna take a long because of are much shorter. Why? Because it’s changes. Now. If we were pulling down the entire database, every time it’s not cost effective and, you know, takes.

[00:42:59] W. Curtis Preston: Because you’re using APIs and you’re doing incremental backup. That makes sense. Cool. Well, listen, uh, we might have lost sanket. It, um, you know, his, his, his bits are flying up in the internet and coming back down all the way from India to here. I, I says he just messaged. He’s not able to hear anymore.

I’m gonna, I’m gonna thank you both for doing this. Sanket can’t hear us anymore, but I’ll, I’ll thank him as well. And, um, and, uh, thanks a lot. I, I, I wish you the best of luck on this new company.

[00:43:28] Vish Reddy: Uh, thank you very much. Uh, uh, and for hosting us out here today, uh, it was great, uh, you know, catching up with you and also sharing what we’ve. Uh, you know, uh, we would love for you guys to try out our product at some point, um, just.

[00:43:44] W. Curtis Preston: Well, well, I I do happen to know a certain, uh, SaaS company that uses JIRA. I’m just saying, and I know some people there and I think you do too, but that’s, that’s a, that’s your own problem to solve, uh, and Prasanna again. Great questions as always.

[00:44:06] Prasanna Malaiyandi: I try and thanks Vish and Sanket. Yeah. Great catching up. Uh, V the one question I was gonna ask. So if people want to try out the product or anything else, I know that the website is going live soon. By the time this podcast gets released, do they just go onto the website, um, to access and they can request a free trial or whatever needs to be done.

Everything’s gonna be available on the.

[00:44:30] Vish Reddy: Great question. Uh, our product is gonna be available on the Atlassian marketplace. Couple of clicks. You can get started. There’s a free trial there. Uh, you know, it’s all automated. Uh, literally you can start your trial of the software within five minutes or maybe less and you get 30 days backup, restore, unlimited, whatever.

Uh, yep. All provision through Atlassian.

[00:44:58] W. Curtis Preston: Nice. I like it. All right. Well, um, again, thanks to our listeners and remember to subscribe so that you can restore it all.


Join the discussion

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: