This generally sounds like something you would want Atlassian's response to - since us community members can be wrong, nor are we willing to be responsible should we indeed turn to be wrong. It's suggestions, at the end of the day.
However, when you delete a project, you delete it - Jira doesn't keep any backups about it. At best you would restore it from native Jira backup (xml) files, if you have those going for any reason, or from an .sql one. So unless you provide those as part of the handover, you should be generally fine.
Issues themselves are stored as 'jiraissue' table - this will get deleted along with the project. This table contains majority of system fields, such as description, due date, summary, reporter, etc.
Project itself is in 'project' table - and it just holds some metadata like name, key, but nothing about data. The row will get removed on delete.
Another table 'customfieldvalue' holds, wait for it, customfield values for the issue. This I believe also gets deleted alongside the project, but not that I ever had the need to confirm it. Plus, the data is linked to the jiraissue table via issue id, so there isn't really any way to "re-construct" the issue, even if the value remained, which I doubt.
Another table is 'jiraaction' which keeps hold of comments. Again, I believe this also does get flushed when the project is being deleted, but never had the need to go in postgres and verify it.
There are 2 conjoined twins 'changegroup' and 'changeitem', which keeps hold of issue history. Again, I believe these to be cleaned when the project is deleted, but never had the need to verify.
There is also something called I think 'issuelink' which keeps sourceissueid:destinationissueid relation about issue link types. This does get deleted also.
These are the major tables that come to my mind. There may be a few more smaller tables such as 'worklog' or 'votehistory' or something similar, but those are pretty much barely even relevant.
What you might want to do though is check for attachments which are normally stored in {{ jira_home }}/data - I believe that the whole project directory gets removed from there along with the project, but it's another thing to confirm it.
All in all, I think if you delete the project, it's pretty much unrecoverable, there's not going to be anything left to reconstruct anything, and Jira generally does a fair job at cleaning it.
What you should be aware of, however, is that the data may persist in some form in 3rd party apps - and this ranges wildly. In majority table from 3rd party apps will be similar in nature, something linked to an issueid, or issuekey, they have no need to keep any of your data, because the data is already there, they just need to link to it. But they could potentially keep some values or information about the issues. Some plugins have cleanup/maintenance jobs to clean themselves, some get automatically cleaned in the database (when a dependent column's value is deleted, for instance), and some clean themselves on the go through event ecosystem. Anyway, it's really a no guarantee about 3rd party apps, they have a lot of freedom in what they store.
You might though want to ping Atlassian for a more "official" confirmation, or at minimum run a few tests with the mentioned db tables, get a few SELECTs going and see what you end up with in the end - but I reckon it should be pretty clean, I don't think you'll find anything to stitch that data back.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.