You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi, I have found out there is some issue with unifi-request node (after updating UDMPro to latest version), which causes high RAM usage after some time (12hrs?).
I have this simple flow where which can be seen on screenshot below. Basicaly it GETs, JSON data from cameras (3x), every 2 seconds (motion detection). I did a bit of research and found out that it seems each call of this node is maybe not being closed - meaning there is a huge amout of opened TCP connections after some time. Please see NETSAT output to get an idea. I think this is root cause.
In the past before UDM update it was allright, but now I had to disable it because it lagged whole NR after some time (high RAM, high CPU).
System info:
RPi4 2GB
UDM Pro v3.1.15
Node-red 2.2.2.
node-red-contrib-unifi-os 0.9.0
Please could you look into it? Thanks & keep up the good work 🥇
The text was updated successfully, but these errors were encountered:
I have gotten a lot of enobufs (WSAENOBUFS) errors in Windows lately. After a lot of searching I did find a command (netstat -ab) to view open connections to find the problem.
When running netstat -ab I did find out that node-red was the problem for all of my enobufs problems. And mostly Unifi connections (and some others).
So I have MANY connections that are not closed to my UMD-Pro
[node.exe] TCP 192.168.1.4:64755 unifi:https ESTABLISHED
After running node-red for 12 hours I have around 2300 open connections in node-red and around 2/3 of them are to unifi. The other 1/3 are to Shelly components.
The command netstat -nao | find /i "estab" /c shows the total sum of connections that are established and that number just keeps growing until I restart node-red.
Hi, I have found out there is some issue with unifi-request node (after updating UDMPro to latest version), which causes high RAM usage after some time (12hrs?).
I have this simple flow where which can be seen on screenshot below. Basicaly it GETs, JSON data from cameras (3x), every 2 seconds (motion detection). I did a bit of research and found out that it seems each call of this node is maybe not being closed - meaning there is a huge amout of opened TCP connections after some time. Please see NETSAT output to get an idea. I think this is root cause.
In the past before UDM update it was allright, but now I had to disable it because it lagged whole NR after some time (high RAM, high CPU).
System info:
RPi4 2GB
UDM Pro v3.1.15
Node-red 2.2.2.
node-red-contrib-unifi-os 0.9.0
Please could you look into it? Thanks & keep up the good work 🥇
The text was updated successfully, but these errors were encountered: