Exception logged #10241
-
Describe the bugmy rabbitmq is v3.8.27,on k8s,I hava enabled rabbitmq_management,After using the plug-in for a period of time, the following error is reported: Ours is a 3-node cluster, and the monitoring plug-in of two nodes reported this error, I don't know what the cause is, can you help solve it? Reproduction steps1.rabbitmqctl set_policy -p zat ha-allqueue "^" '{"ha-mode":"all"}' Expected behaviorHope to know if this is a hidden bug of the monitoring plugin Additional contextno |
Beta Was this translation helpful? Give feedback.
Replies: 2 comments
-
The exception comes from One of the values collected by the plugin from the schema database was FYI, classic mirrored queues are deprecated and will be removed from RabbitMQ in early 2024. Specifically |
Beta Was this translation helpful? Give feedback.
-
Also, RabbitMQ 3.8 has reached EOL in 2022. In fact, 3.10 is about to reach EOL on Dec 31st. |
Beta Was this translation helpful? Give feedback.
The exception comes from
prometheus_mnesia_collector:get_memory_usage/0
, a part of the Prometheus plugin.One of the values collected by the plugin from the schema database was
undefined
. We don't have any context to draw conclusions from but most likely the plugin tried to collect metrics while some tables were still syncing from a peer node.FYI, classic mirrored queues are deprecated and will be removed from RabbitMQ in early 2024.
Specifically
"ha-mode": "all"
has been recommended against for years.