Hey everyone, I'm hitting a major performance wall with our Jira setup and could really use some collective wisdom. 🛠️
We manage a ton of cloud infrastructure projects, and our Jira boards rely heavily on saved filters for dashboards and automation triggers. Lately, some of these filters have become painfully slow—we're talking 30+ seconds to load, and a few time out completely. It's grinding our daily standups and reporting to a halt. The filters often have multiple JQL clauses checking status, team, component, and date ranges across about 50k issues.
Has anyone tackled this successfully? I'm wondering about:
* **JQL Optimization:** Are there specific clauses that are known performance killers? We use a lot of `!=` and `NOT IN` statements.
* **Indexing:** I've heard Jira indexes certain fields. Does adding a certain number of `ORDER BY` clauses break that?
* **Workarounds:** Should we break mega-filters into smaller, cached ones and combine results programmatically via API?
For example, one of our slow filters looks something like this:
```sql
project = INFRA AND status changed DURING (-30d, now()) AND component in (networking, security, kubernetes) AND status not in (Done, Cancelled) AND assignee is not empty ORDER BY updated DESC
```
We're on Jira Data Center, if that makes a difference. Any tuning tips, best practices, or even scripts to analyze filter performance would be a lifesaver!
~CloudOps
Infrastructure as code is the only way