ClickHouse Disk Space Exhaustion

critical

Resource ContentionUpdated Feb 21, 2026

ClickHouse running out of disk space causes NOT_ENOUGH_SPACE errors and prevents trace ingestion. This is a critical infrastructure failure for self-hosted deployments.

Sources

Troubleshooting - Docs by LangChaindocs.langchain.com

Technologies:

LangSmithSymptoms of this issue are visible in LangSmith metrics and logs

ClickHouseThe root cause of this issue originates in ClickHouse

clickhouse.disk.available_bytes

How to detect:

ClickHouse logs show 'DB::Exception: Cannot reserve 1.00 MiB, not enough space: While executing WaitForAsyncInsert. (NOT_ENOUGH_SPACE)'. Monitor ClickHouse disk usage metrics and PVC capacity.

Recommended action:

For Kubernetes: Increase PVC size by editing the storage class to allow volume expansion, patching the PVC, deleting the StatefulSet with --cascade=orphan, and reapplying Helm chart. For Docker: Increase bind mount or volume allocation. Monitor disk usage proactively to prevent recurrence.