Metricbeat RabbitMQ module not sending all metric sets

I am using metric beat 6.3.0 to collect metrics from rabbitMQ. I was able to get it working for the node metricset. However, the connection and queue metric sets do not report any data. I am not seeing any errors in my log file. However, when I run metricbest test modules I do get some errors (next comment). However, it doe snot indicate any way to resolve the error. If you have any ideas I would appreciate it.

Thanks,

Josh

root@localhost:~# metricbeat test modules
...
rabbitmq...
node...OK
result:
{
"@timestamp": "2018-08-14T00:39:25.598Z",
"metricset": {
"host": "127.0.0.1:15672",
"module": "rabbitmq",
"name": "node",
"rtt": 4294
},
"rabbitmq": {
"node": {
"disk": {
"free": {
"bytes": 47495008256,
"limit": {
"bytes": 50000000
}
}
},
"fd": {
"total": 1024,
"used": 31
},
"gc": {
"num": {
"count": 2431137
},
"reclaimed": {
"bytes": 39291174216
}
},
"io": {
"file_handle": {
"open_attempt": {
"avg": {
"ms": 0
},
"count": 9
}
},
"read": {
"avg": {
"ms": 0
},
"bytes": 1,
"count": 1
},
"reopen": {
"count": 1
},
"seek": {
"avg": {
"ms": 0
},
"count": 0
},
"sync": {
"avg": {
"ms": 0
},
"count": 0
},
"write": {
"avg": {
"ms": 0
},
"bytes": 0,
"count": 0
}
},
"mem": {
"limit": {
"bytes": 832916684
},
"used": {
"bytes": 74231808
}
},
"mnesia": {
"disk": {
"tx": {
"count": 0
}
},
"ram": {
"tx": {
"count": 14
}
}
},
"msg": {
"store_read": {
"count": 0
},
"store_write": {
"count": 0
}
},
"name": "rabbit@localhost",
"proc": {
"total": 1048576,
"used": 377
},
"processors": 1,
"queue": {
"index": {
"journal_write": {
"count": 0
},
"read": {
"count": 0
},
"write": {
"count": 0
}
}
},
"run": {
"queue": 1
},
"socket": {
"total": 829,
"used": 0
},
"type": "disc",
"uptime": 280958959
}
}
}

queue...
error... ERROR timeout waiting for an event
connection...
error... ERROR timeout waiting for an event
root@localhost:~#

Hi @joshsmoore,

Indeed the error of metricbeat test modules indicates that metricbeat is not being able to collect any event on these metricsets. This can happen if there is no activity in the cluster and no queues defined, but I guess that you have activity in the cluster. Is it possible that the user you have configured in metricbeat doesn't have access to the vhosts with your queues?

I think it should. Is there an HTTP endpoint I can hit on RabbitMQ to confirm this?

You should be able to access to the rabbitmq management interface with your admin user, there you can check the users and vhosts you have.

Actually you where right about no messages being sent as the problem. It is a staging system and I thought I had sent some messages through but that did not seem to be working. Once I sent some more it is working now.

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.