Thanks for the informations

Yes that's correct athanasios.bladis

Hello Oswaldo Hernandez,

the installed version at BMW is 6.2.9, thats means we need to upgrade our version right?

BR

Athanasios Bladis

Hi athanasios.bladis,

As per my previous comment, I can confirm the issue here was addressed in JIRA 6.4.3.

If you are still seeing a problem in your instance, could you please create a support ticket at https://support.atlassian.com? That way we can provide you with support specific to your circumstances and with higher security. The credentials should be the same as for this site (https://jira.atlassian.com).

Regards,

Oswaldo Hernández.
JIRA Bugmaster.
[Atlassian].

Hi m.hoogenboom,

I can confirm that the issue here was indeed fixed in JIRA 6.4.3 with the upgrade to atlassian-cache library to version 2.5.8.

From your comment I can see that you are seeing a problem whilst creating the project from your plugin's code whilst calling the lower level ProjectManager API. My suspicion is that you might not be clearing all caches that need to be cleaned up / are clearing the cache at the wrong time, but without further context into your code and what you are doing it's difficult to tell what's going on. I would rely on the higher-level ProjectService API instead.

Having said this, it would be better for you to create a question on answers.atlassian.com with more details ad this looks like a developer support question rather than a bug in the product on first sight.

Regards,

Oswaldo Hernández.
JIRA Bugmaster.
[Atlassian].

I happen to see the same issue today in JIRA 6.4.4 with the following code in my jira plugin:
project = projectManager.createProject(projectName, projectKey.toUpperCase(), projectDescription,
leader.getKey(), projectURL, new Long( AssigneeTypes.UNASSIGNED));
}
//... setting schemes according to template ...
projectManager.refresh();

I'm interested in the fix, as it was apparently not fixed in 6.4.4.

Creating a new project (with the default Create Project menu option) resolves the problem, but that's not really an option. I would expect that "projectManager.refresh() would solve it too. Note that I did not have this issue on the Jira 6.4.4 development (atlas-run) environment.

Hello, i m facing the same problem, what was the workaround? Clearing of the Jira internal Cache?

The issue should be fully resolved for JIRA Cloud and JIRA Server customers.

Edit: The deployment to JIRA Cloud had to be delayed due to incompatible code in one of the add-ons that is available there. We expect to be able to address this for the next deployment, but for the time being this error can still occur in JIRA Cloud. See CACHE-123 for more information.

For JIRA Data Center customers, it is still technically possible if two projects are created at once from two different nodes of the cluster. The partial fix implemented here should greatly reduce the occurrences of the problem for JDC customers as well, but a complete, cluster-safe fix is not possible until CACHE-106 and CACHE-108 have been completely resolved.

It is random because it is a race condition. The problem is due to a concurrency problem (see CACHE-106 and CACHE-108 for details). Creating another completely different project works because it resets the two caches that are involved, and the original project that seems to have failed comes along for the ride.

As ohernandez@atlassian.com indicated above, the original fix targeting 6.4.2 had to be reverted due to unacceptable performance impact. I am working on an improved solution that I expect to deliver in 6.4.3, though it is always possible for unforeseen problems to interfere with that plan.

@Jason Brison's most recent edit:
Is that common place? I only experience a "failed" project every ten projects or so and I can't even say that it's that regular. It's more random as I've seen it. Thus creating a new project typically doesn't have any retributions to me.