https://www.umh.app logo
Join Discord
Powered by
Next question, on my "classic" instance
# umh-support
t
Next question, on my "classic" instance, I have the iot sensors that are simulated going to the historian, I can see them i the tag browser. However, any reason why they aren't being published to the MQTT broker in the UNS "umh/v1" topic?
j
they are there for training purposes. so that you can connect the simulators. you can add the default mqtt protocol converter on top of the default mqtt conncetion and it will transfer it to umh.v1
t
so, in my classic instance, I'm not sure I understand how to publish the UNS back into the built in broker....I put in another connection to the broker, with another protocol converter for the sensors....and nothing in the broker is on the topic umh
j
do you see it in the tag browser? what does the protocol converter say how many messages arrive there?
t
Yes, I'm getting them in my historian
yes they are in the tag browser.
But it isn't publishing to the included broker, which is where I'm consuming the iot sensors simulator topics....using the default data flow for mqtt.
j
ok, that is now really strange. if they are of the type _historian, they should also be bridged. I know that in the old bridges there is a lot of confusion about what is bridged at what is not, this is why we are adjusting it at the moment
let me try it out
nvm, my instance has already the new bridges :/
i need to steal an instance from a coleague, mom
t
ok....I have many k3s errors on my classic instance running in a rocky vm after a restart....
maybe it's related? It looks similar to the same k3s errors I'm getting on my "lite" instance running in docker on a pi
j
yes, you are right. in the old bridges data is only bridges by default from MQTT to Kafka, but not the other way around. give us some time to release the bridges :/ if you want to try them out (advanced mode): - scale down the following deployments by accessing the kubernetes cluster https://learn.umh.app/course/how-to-manage-kubernetes-clusters-without-using-kubectl/ databridge-0, mqttkafkabridge - now go to custom data flow components and deploy using the tempalte kafka-to-mqtt-historian-bridge and mqtt-to-kafka-historian-bridge
about k3s errors: can you post them here? in general, kubernetes throws a lot of errors (that are not really errors). we decided to first show them all to the user, but realized that it makes a lot of people insecure and sceptical about our product. so we decided to go through each of the k3s errors and ignore them when the situation is fully known, and/or the error is shown already in other places. during a restart you will get a k3s error about a node restart, but also a lot of events that timescaledb, grafana, hivemq, etc. have all been restarted. k3s events disappear after they have had not happened for some time
t
j
yes, they can be ignored. it means that due to the restart all the microservices were restatyed
t
ok they actually just went away...
the dialog, that is....Thanks...i'll keep that in mind.
j
and they will disappear over time 🙂
and if you have any ideas how we could improve the management console, please tell us! maybe you wish for a tooltip that explains the k3s events? or some tutorials, or a better explanation somewhere give it to us all 😄
t
Either some way to "bury" those errors in another window or button....or an explanation with a tooltip. Or is it possible to just ignore the errors completely, in the console, for the "first hour" which it seems that's when the errors went away. But still have them logged.
2 Views
PreviousNext
Powered by