Issue Summary
We’re experiencing slow text search performance in our ClickStack setup using the external ClickHouse Cloud version.
The schema is the same as in the official Helm deployment guide
.
Observations
Ingress: ~1–3 GB/hour (≈ 300–800 KB/sec)
Egress: ~20–60 MB/hour
Allocated resources: Up to 64 GiB RAM / 20 vCPU
Utilization: CPU and memory usage remain low
Problem: Text search queries are taking longer than expected, despite low resource utilization.
Could the team please review and advise if any tuning recommendations (e.g., indexing strategy, engine settings, or configuration parameters) are suggested for handling large log text search workloads in this ClickStack setup?
syntax_tables.csv