There's a significant difference between a "cache" and an "index". Jira actually has a number of caches, mostly in memory, but you have no direct control over them. Not that you really care - most of them are cleared simply by stopping and restarting it. There is some persistent caching done in application servers, but it's rare that you need to think about them.
The index is a different matter - that's a slab of files on the disk. You really do need that to work properly, because Jira won't really work at all without it.
The good news is that clicking "re index" always fixes the index. This is because it effectively drops the entire thing and rebuilds it from the data in the database from a completely clean start.
The bad news is that if clicking "reindex" doesn't fix your cache, you've propably got quite a serious problem. Your java error may tell us what the problem is.
If you really do want to forcibly clean the index, then stop Jira, delete (recursively) all the files under the "index" directory in "jira home", and restart it and click re-index again. I've never found this fixes anything better than the plain "re-index", but you did ask ;-)
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.