Postmortem: the Tuesday outage
The docs page for this was rewritten alongside it. Existing setups keep working exactly as they did — nothing needs migrating. Two long-standing reports close with this, and both are linked below.
Before you start
Large workspaces should see the biggest difference, since the old query grew with the number of items. Thanks to everyone who filed a report with steps; several of them pointed straight at the cause.
How it works
We will fold the remaining cases in once we have seen this one hold. Anything already in flight when this went out finishes under the old path.
- Enable granular email preferences.
- Enable an audit log.
- Enable SAML SSO.
Step by step
The limit is generous enough that most teams will never meet it. It is slower on the first load and faster on every one after. It works the same way on the public site and inside the widget, with no configuration to turn on.
Limits
It shipped behind a flag first, so a handful of workspaces have been on it for a fortnight. This has been the most requested change on the board for a while, and it is now live for every workspace. The change is backwards compatible, so integrations built against the old shape carry on unchanged.
Common pitfalls
Nothing about billing changes, on any plan. We are watching the numbers for a week before turning it on by default. We rewrote this part rather than patching it, because the old approach ran out of room.
- Enable changelog email previews.
- Enable Slack notifications.
- beta
- breaking