Leap Year Issue: Performance issue

Hi, We are using ElasticSearch 6.2 in production. We noticed a spike in search latency on the recent leap day (02/29/2020).

We use lot of date_range and date fields in our index. We noticed that our search latency almost tripled on the 29th exactly from 00:00:00 (and CPU spiked to 100% as well) and lasted only for the 29th.

Our search rate and index rate was unchanged during the day, and there was no change in query/index fields. This leads us to believe there is something internally in ES which caused this issue.

Is there anything internally in ES that can cause this?

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.