With a lot of new features we released in October, I expected November to be at a much lower pace. But actually I was wrong and in November we released quite a few of interesting features, that any flespi user can benefit from.
Let me start from announcing our monthly uptime — it was 99.9929% in November, with a single 167-second downtime caused by an engineer on site that accidentally blacked out both of our routers at the same time. This should never happen again because the actual reason for that work was to move one of the routers to another rack to provide rack-redundant traffic flow. Now that the works are finished, and we are much more redundant.
At the same time during this move we upgraded our uplink from 1G to 10G so now Enterprise and Ultimate grade clients can enjoy communication with flespi API at multi-gigabit speeds. And we are ready to move to 100G connectivity if need be.
Also, from a datacenter point of view we are now in the process of installing another couple of racks in another Netherlands datacenter and our goal by the end of 2023 is to provide durability of our services on the datacenter level. I mean, if there will be nuclear bomb explosion in one of our datacenters, the whole flespi system will automatically continue to operate from another one. We have a lot of things to implement before we will be able to test this ‘red button’ and completely blackout one datacenter, but the roadmap to this goal is quite clear.
- Once we made billing information and invoices available via REST API and inside flespi.io, we also added the estimated calculation of the next bill for our customers. Now I believe our billing is completely transparent for any user.
- One of the long-awaited features for our users was the possibility to login into flespi.io using non-master tokens. First we allowed to use standard tokens, and recently we enabled login to flespi.io using ACL restricted tokens. It means now our users are able to invite their support teams directly into flespi.io in order to solve some issues. Communication with our team via HelpBox however is available only via master token due to security issues.
- We also implemented the possibility to import and export settings from SetBox. This is a nice feature for those who are using SetBox to configure devices.
- POD IoT modem type was integrated as well. Here is a guide with details and information on how to use modems in flespi in order to perform SMS communication with devices.
- And of course one of the most significant features for developers is our new webhook plugin. This plugin can be used to inject properties or transform the message data somewhere in your backend. Quite a technical thing, but it gives our users significant power for data processing. The lambda for webhook can perform any kind of message modification.
- A few notices about recent changes we’ve done on the platform. First, on December 1 we completely discontinue channel commands. Also, we changed the way device telemetry is published so that only new (by timestamp) telemetry values are being published to MQTT.
- And ongoing update will touch streams subscribed for both channels and devices. Starting December 8 streams subscribed to both device and channel will generate a single message for the device and skip a message for channel (now both are streamed). This is some kind of inconsistency (data duplication) that our users sometimes experience, and we will fix it with this change.
In December, we will continue to add such small and nice features, but our major development flow will be focused on the automation system where our users can define triggers and actions for such triggers. We estimate to release the automation module in some shape in 2023 Q1.
We wish you all a Merry Christmas and a Happy New year! May all the troubles and worries be temporary. Stay safe. Dream big!