mirror of
https://github.com/monitoring-mixins/website.git
synced 2024-12-14 11:37:31 +00:00
44 lines
1.7 KiB
YAML
44 lines
1.7 KiB
YAML
groups:
|
|
- name: ClickHouseAlerts
|
|
rules:
|
|
- alert: ClickHouseReplicationQueueBackingUp
|
|
annotations:
|
|
description: |
|
|
ClickHouse replication tasks are processing slower than expected on {{ $labels.instance }} causing replication queue size to back up at {{ $value }} exceeding the threshold value of 99.
|
|
summary: ClickHouse replica max queue size backing up.
|
|
expr: |
|
|
ClickHouseAsyncMetrics_ReplicasMaxQueueSize > 99
|
|
for: 5m
|
|
keep_firing_for: 5m
|
|
labels:
|
|
severity: warning
|
|
- alert: ClickHouseRejectedInserts
|
|
annotations:
|
|
description: ClickHouse inserts are being rejected on {{ $labels.instance }}
|
|
as items are being inserted faster than ClickHouse is able to merge them.
|
|
summary: ClickHouse has too many rejected inserts.
|
|
expr: ClickHouseProfileEvents_RejectedInserts > 1
|
|
for: 5m
|
|
keep_firing_for: 5m
|
|
labels:
|
|
severity: critical
|
|
- alert: ClickHouseZookeeperSessions
|
|
annotations:
|
|
description: |
|
|
ClickHouse has more than one connection to a Zookeeper on {{ $labels.instance }} which can lead to bugs due to stale reads in Zookeepers consistency model.
|
|
summary: ClickHouse has too many Zookeeper sessions.
|
|
expr: ClickHouseMetrics_ZooKeeperSession > 1
|
|
for: 5m
|
|
keep_firing_for: 5m
|
|
labels:
|
|
severity: critical
|
|
- alert: ClickHouseReplicasInReadOnly
|
|
annotations:
|
|
description: |
|
|
ClickHouse has replicas in a read only state on {{ $labels.instance }} after losing connection to Zookeeper or at startup.
|
|
summary: ClickHouse has too many replicas in read only state.
|
|
expr: ClickHouseMetrics_ReadonlyReplica > 0
|
|
for: 5m
|
|
keep_firing_for: 5m
|
|
labels:
|
|
severity: critical
|