My last half hourly meter reading yesterday was at 19:30 and they resumed at 07:00 today.
Every time this happens, I think to myself "I really ought to be using local access" - but look into it then decide setting it up seems a bit complicated so put it off for a few days then the online ones are back working reliably so forget about it till next time. Has anyone got a straightforward working model of software (preferably in Python or PHP) that I could look at and if necessary adapt ?
KS_Eric, although I believe that it is possible to build an MQTT broker using Python, it is more usual to run something like Mosquitto (a pre-built broker). You will need a system that will run 24x7, but you haven't suggested what platform you might use.
I collect my IHD data locally on an older model Raspberry Pi that runs EmonCMS - I already used it for logging all my solar PV data, and Mosquitto was already installed as part of the standard EmonCMS system. It is, however, quite straightforward to install Mosquitto on any Raspberry Pi system - there are good instructions on the web.
Not really sure what "prometheus metrics" is/are though but hopefully it will make some sense when I can read the code.
At the moment I'm simply downloading "payload.json" from the glowmarkt server then decoding it with a Python script of my own.
And to answer an earlier question, I abandoned my original plan (from around 5 years ago) of running system on a redundant smartphone in favour of using a dedicated 'Lenovo Thinkcentre'.
Prometheus is an open source monitoring and alerting tool (https://prometheus.io/). It takes metrics and stores them in a time-series database. You can then visualise them with Grafana, or other tools. It's an easy way to track the data over time and then left me draw graphs from it.
Comments
Working again now!
Looks like MQTT updates resumed ~ 07:25 BST 😀
Yes (to both points).
My last half hourly meter reading yesterday was at 19:30 and they resumed at 07:00 today.
Every time this happens, I think to myself "I really ought to be using local access" - but look into it then decide setting it up seems a bit complicated so put it off for a few days then the online ones are back working reliably so forget about it till next time. Has anyone got a straightforward working model of software (preferably in Python or PHP) that I could look at and if necessary adapt ?
I'm using a SMETS1 Glowstick and AFAIK local MQTT access is not an option with that, I think that's just on the IHDs?
Thanks DaveS. I should have said : I have the IHD
KS_Eric, although I believe that it is possible to build an MQTT broker using Python, it is more usual to run something like Mosquitto (a pre-built broker). You will need a system that will run 24x7, but you haven't suggested what platform you might use.
I collect my IHD data locally on an older model Raspberry Pi that runs EmonCMS - I already used it for logging all my solar PV data, and Mosquitto was already installed as part of the standard EmonCMS system. It is, however, quite straightforward to install Mosquitto on any Raspberry Pi system - there are good instructions on the web.
Hope this helps
Martin
I created an app that listens to either the local or cloud MQTT broker and exposes the data as prometheus metrics. It's written in Python and you can find it here - https://github.com/andrewjw/glowprom. Connecting to the MQTT broker is handled in https://github.com/andrewjw/glowprom/blob/master/glowprom/mqtt.py, and processing the local messages is https://github.com/andrewjw/glowprom/blob/master/glowprom/local_message.py.
Thanks andrew_j_w ! I'll have a look at that.
Not really sure what "prometheus metrics" is/are though but hopefully it will make some sense when I can read the code.
At the moment I'm simply downloading "payload.json" from the glowmarkt server then decoding it with a Python script of my own.
And to answer an earlier question, I abandoned my original plan (from around 5 years ago) of running system on a redundant smartphone in favour of using a dedicated 'Lenovo Thinkcentre'.
Prometheus is an open source monitoring and alerting tool (https://prometheus.io/). It takes metrics and stores them in a time-series database. You can then visualise them with Grafana, or other tools. It's an easy way to track the data over time and then left me draw graphs from it.