Resolved
Apr 30, 2025 at 8:26pm UTC
We found and patched the root cause of the issue, which was our cache layer. We have added three more cache instances and stopped the processes that were using the cache heavily. We continue to monitor the situation, but the issue should be solved.