Good day. I had
"the flows on the server have been updated" notice, when no update had taken place, combined with an error about a missing node called "Server" which as far as I know I have never installed.
My fix was to look at all my installed nodes in my pallet and update the ones that had a update notification next to them. This required a restart of node-red after which the notices no longer appeared.
I believe it is the equivalent of the above suggestions without using the command line.
I had previously restarted Node-red (power cycle), restarted my browser and PC. With no effect.
I am running Node-red on a Android TV box with Termux.
I've found this happens when I open a window to the editor. Then after awhile, I open a second window (could even be in another browser) and make a change to sometging and deploy. Then I go back to the first window and there is the message. In this image you can see I moved a debug node and deployed and you can see the message on the other window,
After the recent rounds of updates (HA, Node-RED, etc., etc.) I started having this issue with one and only one window open. After a lot of searching, digging, checking, etc., etc. what finally fixed it for me was uninstalling & re-installing the Node-RED addon in HA.
I started having same issue, i am sure i have only one node-red instance running using as a window service (nssm). no browser editor open in any connected computers.
any changes in the only open browser editor, i get a message when i try to deploy that the flow on servers have been updated, and asks me whether to merge .and upon doing (or not doing also) i see that the changes do not appear on the dashboard, i have to restart the node-red service to make effect of the changes. I am confused.
Let me give my two cents on this topic and a fairly different perspective.
I'm running two applications on GCP Cloud Run. They're both pairs - a Dashboard application showing sensory data, and the accompanying admin console - the Node-RED editor responsible for building the Dashboard in every pair.
After deploying the two containers both pairs of applications started to act quirky - the Dashboards refreshed on every 15 to 30 seconds whereas the admin consoles started to show the popups this thread is dedicated to.
Using only one of the apps in the browser didn't change anything ...
What worked in my case? I just increased the container resources:
CPU : 1 > 2
Memory: 512 MiB > 2048 GiB
After the redeployment I never got the same problem again, even when using the for applications in separate tabs simultaneously.
My experience (not with Docker but apps like this in general & VM's): it was most likely the additional RAM that resolved the issue, not the additional CPU.
I bet there is a threshold / magic value that Node needs to operate 'correctly'. I know there are some Node-RED specific options for small-memory systems (Raspberry Pi) that allow overriding memory limits.
I have been using "node-red-contrib-harmony" for years with no issues. About one week ago, I deleted all used nodes in my flow since they were not in use any more and some time later, I got the "The flows on the server have been updated" message quite frequently, just minutes between, and only one browser active. To see if it was a browser problem, I switched from Brave to Firefox but that was exactly the same.
I also noted, in the system log, this message on NodeRed startup:
