If you want to try something on the bleeding edge, you could look into my weewx-mqtt/replicate extension, https://github.com/weewx-mqtt/replicate.
What is weewx-mqtt/replicate? It is my attempt to create a ‘light weight’ way of replicating weewx databases. (I say ‘light weight’ because I have decided that there is no such thing as a light weight replicator.) It uses the request/response functionality of the V5 MQTT protocol. It consists of a service running on a current (the primary) instance of WeeWX and a driver that runs on the secondary instance. Why not just use one of the WeeWX extensions that publishes and another extension to subscribe to the topic? One limitation of this is, if the secondary instance is offline; data will not be replicated. weewx-mqtt/replicate attempts to overcome this. When the secondary instance comes on-line, it asks the service running on the primary instance for all the data since the last record in its database (aka a catchup is performed). Disclaimer 1: It is still very much beta (maybe alpha) code. I have not moved it into production. I only use it to replicate to a development environment. Disclaimer 2: It has limited documentation, so one has to be willing to work through setting it up with limited help. This includes all that goes into setting up a MQTT broker. Disclaimer 3: Not sure if it could replicate multiple databases into a single one. But, if you are interested in experimenting I would be happy to look into supporting this. Disclaimer 4: All of the usual stuff. rich On Saturday, 21 February 2026 at 04:19:32 UTC-5 Tomasz Lewicki wrote: > I have the following idea and problem at the same time. I would like to > build a small network of weather stations in the area, say 5-10 units. Of > course, they will all be powered by Weewx and Raspberry Pi. Apart from the > fact that each one will publish data on the internet independently of the > others (one domain, separate subdomains), I am thinking about a collective > presentation of data from the stations - only selected parameters, e.g., > temperature, wind, precipitation. > > Each Weewx instance is a separate weewx.sdb database with an identical > structure. If I understand correctly, I would have to create a separate > (collective) database in which the indexes of individual parameters would > be different for each station, e.g., temp_pws01, temp_pws02, etc. The idea > is that such a comparison would not be generated in real time, as in a > standard Weewx installation, i.e., every 5-10 minutes, but, say, 4 times a > day. I figured that the databases from each station would be sent via rsync > to a remote server, and then specific columns would be extracted from them > “somehow” and copied to a new, shared database, from which a summary would > be made. Can the final step be done with Weewx, or do I need to use other > tools? > > The question is, has anyone done something like this before and could > point me in the right direction? > > *Translated with DeepL.com > <https://www.deepl.com/?utm_campaign=product&utm_source=web_translator&utm_medium=web&utm_content=copy_free_translation> > > (free version)* > -- You received this message because you are subscribed to the Google Groups "weewx-user" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion visit https://groups.google.com/d/msgid/weewx-user/66503425-b476-412d-b48a-91d378110dbfn%40googlegroups.com.
