Resolved -
A significant increase in downlink queue operations within a short time window and this caused internal database connections to accumulate as each request waited for exclusive access to device resources. This connection buildup exhausted the available connection pool, resulting in degraded performance for other operations across the Application Server.
Unrelated requests experienced increased latency and timeouts as they competed for the limited available connections. Measures were taken so that this incident do not happen again.
Dec 19, 17:53 CET
Monitoring -
A fix has been implemented and we are monitoring the results.
Dec 19, 16:26 CET
Investigating -
We are investigating reports about webhook failures in our eu1 cluster.
Dec 19, 15:50 CET