Sweethawk 장애 기록

App server (legacy) went down

App server (legacy) went down

There were jobs taking excessive resources. We have fixed the performance of these jobs and service has been returned to normal.

App server (legacy) went down

We have now rebuilt our cache server cluster and are bringing things back online. We will continue to monitor but if you notice anything out of the ordinary please email support@sweethawk.com.

Apps are currently not working due to our Redis server being unavailable. We are working on the issue. Apologies for the service interruption.

App server (new platform) went down

Recurring Tickets are working again after an interruption. Some tickets may have been created later than their schedule by up to 24 hours later. We have also addressed an issue with the Timers app, having a similar issue. Some timers that were started over 24 hours ago and were due to complete in the last 24 hours may have been missed. If you have any concerns, please contact us at support@sweethawk.com, kindly let us know example ticket ID(s) we can look into and that we have access to your data for troubleshooting purposes and we'll get back to you asap.

Short downtime is required to do a database system update.

The issue has been resolved, all apps working again as normal.

Due to a database issue, the Notify app is currently degraded, we are unable to surface whether there are unread notifications when the app loads. Apologies for the trouble, we hope to have this resolved soon.

The apps are working again and we have rolled back a recent release. We are looking into the cause.

We are having issues with our database on our legacy app server and are looking into the cause, apologies for the outage.

We are looking into an unexpected outage caused by a database schema upgrade. Apologies for the service interruption.

An issue with jobs (eg approvals, task lists, timers) not processing in a timely manner has been resolved. Due to a temporary issue with a response from the Zendesk job status this also caused an issue on our side, some jobs were running too long and causing a backlog in the processing queue. The issue was temporary and we have made a fix that prevented some other jobs from waiting too long on others to expedite clearing the queue.

Maintenance completed

The apps will become offline for a short period of time.

Planned maintenance to the database has been completed successfully.