Skip to content

Commit 208c1d8

Browse files
authored
Merge pull request #709 from stackhpc/rabbitmq-alert-fixes
Improve RabbitMQ alerts
2 parents cfd80e1 + c6fc057 commit 208c1d8

File tree

2 files changed

+32
-20
lines changed

2 files changed

+32
-20
lines changed

etc/kayobe/kolla/config/prometheus/rabbitmq.rules

Lines changed: 27 additions & 20 deletions
Original file line numberDiff line numberDiff line change
@@ -5,69 +5,76 @@
55
groups:
66
- name: rabbitmq.rules
77
rules:
8-
- alert: RabbitmqNodeDown
9-
expr: sum(rabbitmq_build_info{host_name!=""}) < 3
8+
- alert: RabbitMQNodeDown
9+
expr: sum(rabbitmq_build_info{instance!=""}) < 3
1010
for: 30m
1111
labels:
1212
severity: critical
1313
annotations:
14-
description: Rabbitmq node down on {{ $labels.host_name }}
15-
- alert: RabbitmqConsumersLowUtilization
14+
description: RabbitMQ node down on {{ $labels.instance }}
15+
- alert: RabbitMQConsumersLowUtilization
1616
expr: rabbitmq_queue_consumer_utilisation < 0.4
1717
for: 5m
1818
labels:
1919
severity: warning
2020
annotations:
21-
description: RabbitMQ consumers message consumption speed is low on {{ $labels.host_name }}
22-
- alert: RabbitmqNodeNotDistributed
21+
description: RabbitMQ consumers message consumption speed is low on {{ $labels.instance }}
22+
- alert: RabbitMQNodeNotDistributed
2323
expr: erlang_vm_dist_node_state < 3
2424
for: 5m
2525
labels:
2626
severity: critical
2727
annotations:
28-
description: Rabbitmq node not distributed on node {{ $labels.host_name }}
29-
- alert: RabbitmqMemoryHigh
28+
description: RabbitMQ node not distributed on node {{ $labels.instance }}
29+
- alert: RabbitMQMemoryHigh
3030
expr: rabbitmq_process_resident_memory_bytes / rabbitmq_resident_memory_limit_bytes
3131
* 100 > 90
3232
for: 2m
3333
labels:
3434
severity: warning
3535
annotations:
36-
description: Rabbitmq memory too high on {{ $labels.host_name }}
37-
- alert: RabbitmqFileDescriptorsUsage
36+
description: RabbitMQ memory too high on {{ $labels.instance }}
37+
- alert: RabbitMQFileDescriptorsUsage
3838
expr: rabbitmq_process_open_fds / rabbitmq_process_max_fds * 100 > 90
3939
for: 2m
4040
labels:
4141
severity: warning
4242
annotations:
43-
description: Rabbitmq file descriptors usage on {{ $labels.host_name }}
44-
- alert: RabbitmqTooMuchUnack
45-
expr: sum(rabbitmq_queue_messages_unacked) BY (queue) > 1000
43+
description: RabbitMQ file descriptors usage on {{ $labels.instance }}
44+
- alert: RabbitMQTooMuchReady
45+
expr: rabbitmq_queue_messages_ready > 100
4646
for: 1m
4747
labels:
4848
severity: warning
4949
annotations:
50-
description: Rabbitmq too much unack on {{ $labels.host_name }}
51-
- alert: RabbitmqTooMuchConnections
50+
description: RabbitMQ too much ready on {{ $labels.instance }}
51+
- alert: RabbitMQTooMuchUnack
52+
expr: rabbitmq_queue_messages_unacked > 100
53+
for: 1m
54+
labels:
55+
severity: warning
56+
annotations:
57+
description: RabbitMQ too much unack on {{ $labels.instance }}
58+
- alert: RabbitMQTooMuchConnections
5259
expr: rabbitmq_connections > 1000
5360
for: 2m
5461
labels:
5562
severity: warning
5663
annotations:
57-
description: Rabbitmq too much connections on {{ $labels.host_name }}
58-
- alert: RabbitmqNoQueueConsumer
64+
description: RabbitMQ too much connections on {{ $labels.instance }}
65+
- alert: RabbitMQNoQueueConsumer
5966
expr: rabbitmq_queue_consumers < 1
6067
for: 1m
6168
labels:
6269
severity: warning
6370
annotations:
64-
description: Rabbitmq no queue consumer on {{ $labels.host_name }}
65-
- alert: RabbitmqUnroutableMessages
71+
description: RabbitMQ no queue consumer on {{ $labels.instance }}
72+
- alert: RabbitMQUnroutableMessages
6673
expr: increase(rabbitmq_channel_messages_unroutable_returned_total[1m]) > 0 or increase(rabbitmq_channel_messages_unroutable_dropped_total[1m]) > 0
6774
for: 2m
6875
labels:
6976
severity: warning
7077
annotations:
71-
description: Rabbitmq unroutable messages on {{ $labels.host_name }}
78+
description: RabbitMQ unroutable messages on {{ $labels.instance }}
7279

7380
{% endraw %}
Lines changed: 5 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,5 @@
1+
---
2+
features:
3+
- |
4+
Raises an alert when the count of RabbitMQ ready messages increases above a
5+
threshold.

0 commit comments

Comments
 (0)