[Questions] Occasionally, after deploying RabbitMQ, the pod status of openstack-rabbitmq-rabbitmq-1 is CrashLoopBackOff. rm /var/lib/rabbitmq/ and then restarting the pod resolves the issue. #13562
Replies: 1 comment
-
@sheshesheshe you were told earlier that RabbitMQ 3.12 is out of community support. It will be out of support even for paying customers in just a few months. In the few cases where Mnesia failed to merge its schema, there were different factors in play, We highly recommend using our own cluster Operators on K8S, or at least learning from them and the docs. We won't investigate an issue with a specific 3rd party Helm chart. Upgrading to 4.0.x and Khepri will fundamentally solve this because Mnesia simply won't be used in such a cluster. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Community Support Policy
RabbitMQ version used
other (please specify)
Erlang version used
26.0.x
Operating system (distribution) used
k8s pod,rabbitmq: 3.12.14-management
How is RabbitMQ deployed?
Community Docker image
rabbitmq-diagnostics status output
Because the RabbitMQ pod status is CrashLoopBackOff, I am unable to execute the rabbitmq-diagnostics command.
Logs from node 1 (with sensitive values edited out)
Logs from node 2 (if applicable, with sensitive values edited out)
See https://www.rabbitmq.com/docs/logging to learn how to collect logs
Logs from node 3 (if applicable, with sensitive values edited out)
See https://www.rabbitmq.com/docs/logging to learn how to collect logs
rabbitmq.conf
enabled_plugins: |2
rabbitmq-env.conf: |
SERVER_ADDITIONAL_ERL_ARGS="+S 1:1"
rabbitmq.conf: |
cluster_formation.k8s.address_type = hostname
cluster_formation.k8s.host = kubernetes.default.svc.cluster.local
cluster_formation.node_cleanup.interval = 10
cluster_formation.node_cleanup.only_log_warning = true
cluster_formation.peer_discovery_backend = rabbit_peer_discovery_k8s
cluster_partition_handling = autoheal
listeners.ssl.1 = 5672
listeners.tcp = none
log.console = true
loopback_users.guest = false
management.ssl.cacertfile = /etc/rabbitmq/certs/ca.crt
management.ssl.certfile = /etc/rabbitmq/certs/tls.crt
management.ssl.keyfile = /etc/rabbitmq/certs/tls.key
management.load_definitions = /var/lib/rabbitmq/definitions.json
management.ssl.port = 15680
queue_master_locator = min-masters
ssl_options.cacertfile = /etc/rabbitmq/certs/ca.crt
ssl_options.certfile = /etc/rabbitmq/certs/tls.crt
ssl_options.fail_if_no_peer_cert = true
ssl_options.keyfile = /etc/rabbitmq/certs/tls.key
ssl_options.verify = verify_peer
Steps to deploy RabbitMQ cluster
chart:https://opendev.org/openstack/openstack-helm-infra/src/branch/master/rabbitmq
helm upgrade openstack-rabbitmq openstack/rabbitmq --install --create-namespace --namespace openstack --version 3.12.14 --set pod.replicas.server=3
Steps to reproduce the behavior in question
I have a 3 node RabbitMQ cluster. All queues are quorum type.

Occasionally, after deploying RabbitMQ, the following command is used:
helm upgrade openstack-rabbitmq openstack/rabbitmq --install --create-namespace --namespace openstack --version 3.12.14 --set pod.replicas.server=3
advanced.config
See https://www.rabbitmq.com/docs/configure#config-location to learn how to find advanced.config file location
Application code
# PASTE CODE HERE, BETWEEN BACKTICKS
Kubernetes deployment file
What problem are you trying to solve?
I have a 3 node RabbitMQ cluster. All queues are quorum type.

Occasionally, after deploying RabbitMQ, the following command is used:
helm upgrade openstack-rabbitmq openstack/rabbitmq --install --create-namespace --namespace openstack --version 3.12.14 --set pod.replicas.server=3
The log file
rabbitmq-1.log
indicates the issue as described. I want to know the reasons for this issue, whether it has been fixed in newer versions of RabbitMQ, and if so, which commit this issue. Additionally, I would like to know if there are any workarounds to avoid this problem.Beta Was this translation helpful? Give feedback.
All reactions