At 13:06 Pacific Apollo underwent a production outage for our Studio UI features. Both internally and externally, users logged into Studio were being logged back out after 30 seconds. This outage ended at 13:44.
This was due to a production deploy of some new feature flagging logic, which enables some Enterprise logout functionality. Instead of this functionality being localized, and enabled for testing, we ended up turning this on for everyone. At 13:44, the issue was identified and we promptly disabled the errant feature flag.
We understand the kind of disruption this caused for everyone (not least of which because at Apollo we also dog food our own product and were getting booted after 30 seconds). We will be conducting our post mortem for this in the coming days as we continue to strive to provide you with the best possible service.
As always, we appreciate your patience!
No components marked as affected
Resolved
At 13:06 Pacific Apollo underwent a production outage for our Studio UI features. Both internally and externally, users logged into Studio were being logged back out after 30 seconds. This outage ended at 13:44.
This was due to a production deploy of some new feature flagging logic, which enables some Enterprise logout functionality. Instead of this functionality being localized, and enabled for testing, we ended up turning this on for everyone. At 13:44, the issue was identified and we promptly disabled the errant feature flag.
We understand the kind of disruption this caused for everyone (not least of which because at Apollo we also dog food our own product and were getting booted after 30 seconds). We will be conducting our post mortem for this in the coming days as we continue to strive to provide you with the best possible service.
As always, we appreciate your patience!