Hi all,
Looking for ideas.
I updated some nodes from the pallette manager this afternoon, and now NR is running very slowly. Take and age for the UI to load, Deploys take over 10 secs (used to be <2s) and getting a number of repeating entries in the docker log for the container. Error connecting to dbus: Error: connect ENOENT /var/run/dbus/system_bus_socket
is the repeating entry.
One of the nodes updated was home-assistant-websocket - it needed to auto update a large number of nodes, and now almost all HA nodes either don't work, or will work but with a very long delay.
I've restarted both NR running in a docker container, and Home Assistant running on a Pi4.
No other changes made, so thinking it's the HA node with the problem, but not seeing any other comments on it.
Option to uninstall home-assistant-websocket is not visible in the pallette manager.
Thanks
Also worth mentioning that the memory use for the NR container has doubled and the CPU usage is very high. (Might explain the long delays loading and updating).
I think that may mean that something in your container is trying to use dbus but the container has not be setup to allow that. I don't know what would be using dbus though, unless it is is a HA issue. Have you tried asking on an HA forum? Few here use HA.
The high CPU might be a result of the dbus problem, or perhaps you have some sort of loop in your system. If you definitely have not changed the flows then that does not seem likely.
It was super stable before updating node, unfortunately I updated about 6 node sets at the same time so not really sure which one upset the applecart...
I have snapshot backups of the /data volume so I guess I'll just do a restore, but that feels like defeat... lol
Restore it and then update them one at a time (if the restore does fix it), it may not be related to that update, tricky problems are often the result of strange co-incidences.
