Transparency and preventing nasty surprises: Operation packages for Atlassian applications with monitoring

Pamela Winter awakes from a dreamless sleep. She looks at the clock. Plenty of time until her meeting at 9:30am.

Sitting at the kitchen table with a cup of coffee, she picks up her phone and goes for her first quick news fix of the day: Twitter. She'll look at her emails when she gets to the office. They're no longer as important in her company anyway. Not since they've been working with Atlassian software.

While she sips her coffee, she scrolls through the latest tweets, likes one or two of them – and as usual half way down her timeline Twitter says "You might like" – followed by a list of recommended accounts generated by Twitter's algorithms.

"Well, I never! Theo Schroeder? I think I actually know him. Is he a colleague from work?" Pamela clicks on his profile. Yes, that's right. "I sat with Theo at lunch last week. What does he tweet about? Is it worth following him?"

Pamela suddenly looks up from her phone. She has to be careful not to spit out a mouthful of coffee. Then she looks again. This tweet really is from this morning. That can't be for real!

She frantically looks at her watch: it's almost 8:30am. What the hell is going on over there?! Have the teams been sitting around without access to one to their most important systems for nearly 90 minutes, and she – Pamela, the IT manager – was none the wiser? Is one of her team already on it?

What now? Well, first of all, she needs to get to the office! Ideas for swift action run through her head: Call her team members and round them up! And what about the software provider – has someone already called them and put on the pressure?

"Damn it, that idiot Steven Scott is probably knocking at management's door to show them his calculations for every single minute of lost productivity, down to the last cent. Oh, man..."

Pamela's morning is shot. Now she has to get the team together and do everything they can to get Jira back up and running – people at the office need it. What a mess!

Pamela awakes from a strange dream. She looks at the clock. Plenty of time until her meeting at 9:30am.

She's probably worrying about nothing. But dreaming about work is certainly not a good thing. Perhaps she should go away for a week so that she can clear her head and then return to work rested and refreshed.

What was she dreaming about anyway? The dream is already beginning to fade, but she still remembers vaguely: Jira was down, everything had come to a standstill. Noone knew what was going on, Pamela had only found out about the problem by pure coincidence. Wow, that would be all she needs! That'd be just what Steven Scott has been waiting for, that idiot.

She sits at her kitchen table, chuckles to herself and drinks a sip of her coffee, lost in thought. Good job, such a mishap wouldn't happen to us. Of course, Jira can go down sometimes. IT systems are complex. You can minimize problems but never eliminate them entirely. You can take all the precautions you like – and then all it takes is a small typo to set off a chain reaction that puts half of the cloud out of action.

When there are technical problems with an Atlassian system at Pamela's company, there are tools and people that know what to do. Their Jira has an extensive operation package delivered by their hosting provider, which includes continuous surveillance: A central monitoring system keeps watch over the application around the clock. The SLA determines the exact scope of the service and its timeframes in line with the enterprise's specific needs.

Here are the components and metrics that the monitoring system checks and logs:

  • Web server – application availability, simultaneous connections, SSL certificate's expiry date
  • Atlassian – Java Heap memory, threads, and open files
  • Data security – local TFS snapshots, mirroring to backup server
  • System – CPU utilization, disk space, internet access, server time, email dispatch

Some metrics are checked every 10 seconds so that the timeline and development of the issue and metrics are always ready for the problem-solving process. Below is an example of typically monitored data:

Since this level of transparency and security was introduced, fast reactions to problems and issues is guaranteed. As soon as a problem is detected, the system administrators at the service provider are informed of critical cases (within the service time outlined in the SLA) and independently start working on correcting it.

And of course, the customer is also informed – in this case, it would be Pamela's team. So, Pamela would know before she came into the office at 9:30am or found out about it by coincidence. She herself or a contact person from her team would be brought into the picture quickly and maintain contact with the administrator team throughout.

Pamela shakes the thought and opens Twitter on her phone. Let's see what's going on in the world. She scrolls down her timeline, and Twitter suggests some people she might want to follow. Oh, Theo Schroeder – that name seems familiar and the picture too. Didn't I sit with him at lunch just last week?

But there's something else. Pamela is briefly overcoming with a feeling déjà-vu, as though she had just spoken to Theo

Be that as it may. Without further ado, she heads to Theo's profile and reads his latest tweet:

Theo Schroeder tweets

Oh no, the poor guy, Pamela thinks to herself. There's always something going on. But she's sure someone in the office management team is already onto getting the coffee machine fixed. And she's relieved that it's the coffee machine and not Jira. That would be worse for Pamela, her team, and the company.

Lesen Sie diese Seite auf Deutsch

Your partner for all your Atlassian software operations

//SEIBERT/MEDIA offers attractive operation packages to run your Atlassian applications securely and reliably. Our packages can be tailored to your specific needs with SLAs, continuous monitoring, and systematic backup concepts. We are an Atlassian Platinum Solution Partner: Our experienced technical teams are experts on the particularities of Atlassian systems and have created operation and hosting packages based on these specific requirements.

Do you want to find out more? Our Knowledge Base contains detailed information on our system administrative Atlassian services. Do you have any questions? Then feel free to get in touch! We'd be happy to help!