I encountered a problem that my NodeRED agents began to disconnect, going into the “missing” status. This began to happen recently when the number of agents increased from 3 to 8 in one day. I couldn’t track how long they work, but it feels like they work for 1-2 days and then turn off
At first I thought that the Linux server did not have enough resources. I added RAM and allocated 2 cores for the processor, but the situation has not changed
Any agent can stop working, both old and new ones that I added recently. The key here is that this started happening when I added 5 new agents, but I’m not sure if the problem is directly related to this
But it was not possible to get logs of fallen containers, since they are not displayed at all in the list.
Then they where deleted, not stopped.
Most likely that is openflow then, does it happen every 8 hour by any chance ?
If so, it could be the housekeeping job, that among many things also removed agents that has been running for more than 8 hours and does not have the billing tag.
This feature is enabled if you have enabled multi tenancy and/or have created a “agent” resource ( so you can select between free and other sizes when creating an agent )
Yes, I created this resource on February 12th. And on the same day problems began with the agents. I created a resource using the ensurecommon button (I was just studying the possibilities of OpenFlow in my free time).
I deleted the resource and when creating an agent I can no longer choose a paid plan. I’ll be monitoring the agents’ behavior for a couple of days. I will post the results in this topic
No, create the resources, and then assign a plan to all agents.
( you need to select a plan and click save, and then stop/start the agent to “tag” it )