Very disapponted by the server policy change

[qbprog]

Just to you know, I'm very disappointed by the deprecation of the server/on-premises products.

We will not ever move our repositories to the cloud, so we will just look for an alternative when supports end.

[Robert Giesecke]

Yeah, apparently we just wasted a lot of time and money finding a company to manage our on-premise servers. And I’ll have to figure out whether to keep investing in that endeavor or not...

We can’t go cloud where all plugins have their own storage all over the web, we can’t audit every one of those regularly. So using it would mean a constant audit finding for us.

What would even be an alternative for confluence?
Bitbucket server has gitlab, Jira has enough rivals. Hopefully one has something like JQL.

[Timo Reitschmidt]

I just received this announcement via mail and I am totally shocked. We are currently evaluating Bitbucket and Jira because of the on premise option and we are not allowed to store anything in the cloud. I will put the project on hold on Monday when I am back in the office. This is ridiculous to go cloud only. The complete invest was already approved and now we have to restart the whole project.

Cheers, Timo

[Julius Davies [bit-booster.com]]

The DC versions will remain available.  It's only the Server versions that are going away.  Of course DC license renewals are 2x the price of Server renewals, so it does double your licensing costs after Feb 2024, but nonetheless, on-prem remains an option.

[roman_baeriswyl]

I‘m very disappointed. I do not want or even can have that data in the cloud. Also what about offline scenarios? There are some project environments that are strictly offline.

If they stay with their roadmap, they will loose a lot of customers including myself and probably also my company to other products. They will also loose plugin developers and plugin maintenance as it will be not worth the development time anymore.

[a_learner]

Who thought up this crazy idea?  I think this is a "lets shoot ourselves in the foot" strategy.   I have worked in large corporates and there is no way they will put their private source code in the cloud.  These are companies that spend hundreds of thousands or millions on the right tooling and security.   Lets see how this plays out.  Bitbucket is not the only on premise game in town.  Bye bye Bitbucket.

[Julius Davies [bit-booster.com]]

They are not shooting themselves in the foot.  But they are increasing the cost of on-prem by 2x !   (DC license renewals are twice as expensive as Server renewals).

[Ulrich Kuhnhardt _IzymesCo_]

The Datacenter product versions remain as an on-premise option. As far as I know most Atlassian products (except Bamboo) offer a DC version for on-prem deployment. 

[Julius Davies [bit-booster.com]]

And the EOL for all of these "Server" products (assuming you don't want to upgrade to DC) is February 2, 2024 !    3.25 years from now.  

(That's when they stop receiving bug fixes and security patches).

[roman_baeriswyl]

Who guarantees that they won't shutdown the DataCenter versions in 3 years? I think it is very likely that they shut them down later as well as they are going full cloud. So at latest then we have to migrate to something else, so why first migrate to DC at very high costs and afterwards to something else?

[Alex Janes]

There are no guarantees on the Data Center version. Per their earnings report, it sounds like there is an unannounced EoL for DC, as they say all of their customers (including data center) will move to the cloud in the medium term.

See more at this post...

https://community.atlassian.com/t5/Feedback-Forum-questions/Re-Re-Answering-your-questions-on-changes-to-server-and/qaq-p/1522838/comment-id/2454#M2454

[Ulrich Kuhnhardt _IzymesCo_]

Here is an interesting article (sorry in german, but you can pipe it through google translate probably) on how Atlassian's all-cloud-all-in strategy is perceived from the EU - https://www.heise.de/news/Wie-sicher-sind-Geschaeftsdaten-nach-der-Atlassian-Migration-in-die-Cloud-5048772.html

[qbprog]

nice article!

not sure if their expectation will become reality, they surely did their math; eventually moving tickets to the cloud could be acceptable, but for code ? noooo way.