Skip to:
In the logs posted on , we see newer config versions being processed, but retries going against the old configs.
When the reason is NotMyVBucket, these retries are destined to fail.
This happens when
A multi-node cluster is scaling up (adding nodes)
The system is under constant load
The latency is high
Fixed by
In the logs posted on , we see newer config versions being processed, but retries going against the old configs.
When the reason is NotMyVBucket, these retries are destined to fail.
This happens when
A multi-node cluster is scaling up (adding nodes)
The system is under constant load
The latency is high