Web UI inexplicably indicating very low in/out msg/s with errors being logged in server log and system messages repeatedly.
Expected Behavior
When the UI was displaying correctly, it would typically show 7k-25k msg/s constantly. There is no possibility that we are sending that few messages because all our environments are logging to this cluster.
Current Behavior
The In/Out msg/s in web UI shows 0/0 most of the time. It occasionally ticks up but goes back to 0/0. The actual amount of messages being logged seems to be correct. We are also seeing a high number of entries in the server log showing "org.graylog2.periodical.NodePingThread - Did not find meta info of this node. Re-registering." System messages in web UI also shows "Notification condition [NO_MASTER] has been fixed."
Steps to Reproduce (for bugs)
Started Graylog and it soon goes into this state.
Context
I am not entirely sure if this is a display problem or is actually causing messages to be dropped.
Your Environment
We have a Graylog collector server (master), and a Graylog web UI server (slave), a MongoDB server and 5 ES data nodes. All log messages are being sent to the collector server, and users access Graylog via the web UI server.
Web UI inexplicably indicating very low in/out msg/s with errors being logged in server log and system messages repeatedly.
Expected Behavior
When the UI was displaying correctly, it would typically show 7k-25k msg/s constantly. There is no possibility that we are sending that few messages because all our environments are logging to this cluster.
Current Behavior
The In/Out msg/s in web UI shows 0/0 most of the time. It occasionally ticks up but goes back to 0/0. The actual amount of messages being logged seems to be correct. We are also seeing a high number of entries in the server log showing "org.graylog2.periodical.NodePingThread - Did not find meta info of this node. Re-registering." System messages in web UI also shows "Notification condition [NO_MASTER] has been fixed."
Steps to Reproduce (for bugs)
Started Graylog and it soon goes into this state.
Context
I am not entirely sure if this is a display problem or is actually causing messages to be dropped.
Your Environment
We have a Graylog collector server (master), and a Graylog web UI server (slave), a MongoDB server and 5 ES data nodes. All log messages are being sent to the collector server, and users access Graylog via the web UI server.