How long is each ClickHouse release supported?
ClickHouse, Inc. supports the current stable release and the two before it for at least three months, and the current and previous long-term support (LTS) release for at least a year. LTS releases are typically March and August. A cluster not upgraded for a year is probably outside vendor support, and AceMQ supports any release you run.
Who answers when ClickHouse ingest stops, and how fast?
A named AceMQ senior engineer answers within 15 minutes, 24/7, when inserts are failing or queries are timing out cluster-wide. Lower severities get a first response within 1 hour, 4 hours or the next business day. The engineer who takes the call stays on the case, and you get a written root-cause analysis after every P1.
How does a ClickHouse engagement with AceMQ start?
It starts with a free consultation with a senior engineer, booked in our reply within 1 hour. A table and ingestion review of 1 to 2 weeks follows, covering sorting keys, partitioning, part counts, insert batch sizes, materialized views and shard distribution. Written findings and a plan arrive within a week, and nothing changes before you have read them.
Can an AceMQ engineer work alongside our in-house team?
Yes. A named senior engineer works with your engineers rather than around them, changing table design and producer batching together. New table structures are populated in parallel and swapped in behind the existing name, so queries keep serving, and you keep the runbooks for part-count alerts, replica recovery and schema change.
What is the too many parts error, and how is it fixed?
Too many parts means background merges cannot keep up with new parts, almost always because inserts are too small or too frequent. Fix it at the producer by batching tens of thousands of rows per insert or enabling async_insert, and check that the partition key is not too granular. Raising the parts limit only trades an error today for a slower cluster later.
Why does a filtered ClickHouse query still scan the whole table?
ClickHouse has a sparse primary index over the ORDER BY key, so a filter only prunes data if it matches a prefix of that key. Filtering on a later column alone reads everything. AceMQ measures whether a skip index, a projection or a rebuilt table with the right sorting key wins on your data, instead of guessing.
Should we use ClickHouse Cloud or run ClickHouse ourselves?
ClickHouse Cloud removes Keeper, replication and storage provisioning, which is a real saving. It does not remove schema design, sorting keys, mutation cost or insert batching, which cause most incidents either way. Self-managed suits any version, any host and air-gapped estates. AceMQ supports both, and can operate self-managed clusters in your own account.
Can AceMQ migrate us from Elasticsearch or Druid to ClickHouse?
Yes, where the workload is aggregation heavy rather than full-text search. AceMQ benchmarks ClickHouse against your real queries first, converts the schema, runs dual writes to compare results and backfills history before cutover. If the benchmark favours staying where you are, we say so.
Can we work with AceMQ on ClickHouse without a long contract?
Yes. AceMQ works on ClickHouse in two ways: ad hoc engagements for a defined piece of work, such as a health check, an upgrade or an incident, or an annual partner engagement that covers 24/7 support, proactive reviews and named engineers. Many teams start ad hoc and move to an annual partnership once the first engagement is done.
How much does AceMQ ClickHouse support or consulting cost?
Every AceMQ proposal is tailored to your requirements: your ClickHouse environment, the SLA you need, the hours of cover and the scope of work. Most contracts run $24,000–72,000 a year, depending on the SLA, the scope and how ClickHouse is deployed, with aggressive discounts when several technologies share one contract. Request a quote and AceMQ replies within 1 hour.
Kafka
Kubernetes
Prometheus
Grafana
Linux
S3
