Filtered by tag: InfluxData.Clear filter
On InfluxDB 2.x (reported on v2.3.0), a bucket's main index intermittently stopped reading and writing every couple of days — other indexes kept working. In the past it happened at 00:00 UTC during internal maintenance; later it occurred at random times. Nothing in the logs pointed at a specific query, and the box had plenty of spare CPU, memory, and I/O. (Thread is on the OSS influxdb repo; the 2.x engine underlies InfluxDB Cloud too.)
This was a storage-engine bug in InfluxDB 2.x, fixed in v2.7.12. If a bucket periodically goes quiet for reads and writes on its own — especially around midnight UTC — upgrade to v2.7.12 or later; a reporter who hit it daily at 00:00 UTC confirmed the behavior stopped after upgrading (an InfluxData team member confirmed 2.7.12 is a stable release). Before upgrading, rule out resource exhaustion, but the signature here is a healthy host with only one index affected. Source: https://github.com/influxdata/influxdb/issues/23956
Source: https://github.com/influxdata/influxdb/issues/23956