I have convinced my client to move from MS SQL to InfluxDb for the project I am working on. This client now pays 100s per month for the cloud offering. As I am building out the project I stumble on “Flux now in maintenance mode” and the news it’s not supported towards the future. All our backend code retrieves Influx data using Flux through the query REST api. And I had just managed to convince one of the engineers at the client to invest some time and learn this so he could do his own analysis. He was thrilled. But now I’m stuck and frustrated.
If InfluxQL is the future then why is it not a first class citizen on v2? When I open the site I can only query data using Flux. So no one can develop skills in this and I can’t change backend code yet in anticipation of v3. I read we could use it in v2 but you need table mappings or something. But that sounds like a hack and doesn’t convince me that it’s mature on v2.
My point is: in a professional environment you cannot expect me to lead them into a transition from v2 to v3 when BOTH the DB backend will be upgraded AND I need to change the business logic at the same time. I need an overlap. Either InfluxQL needs to work as a first class citizen on v2 or Flux needs to work on v3 with maybe only a change to the REST url.
And with this move to v3: it is entirely unclear to me what the strategy is for paying customers on the cloud offering. Is there a date when everything will move from v2 to v3? Will there be a (web) UI just like there is now? On Youtube I find videos from a year ago where v3 is presented in short clips telling me it’s the future. But no details about the migration for customers.
For paying customers, we’ll continue to operate and support previous offerings. We appreciate that the different versions and API changes can be difficult, which is why we continue to run previous version for our customers years after release.
We don’t yet have the migration pathways and tooling to get to v3 from previous versions. For some customers we can handle this on a case by case basis.
We’ll be working on this tooling and the overall user experience over the course of the next year. In the meantime, if you’re a Cloud2 customer, you can continue to use it without interruption.
OK. So InfluxDb will remain as a set of products with their own development paths I gather. I note this is somewhat unusual in the software industry.
What impact does having InfluxDb3 have on InfluxDb2 future development? I’m guessing InfluxDb2 is now in “maintenance mode” where you only do security/stability fixes. Correct? But no new features of UI revamps I’m guessing.
Is this really the end of the line for Flux? Will there be a replacement?
Regarding 2. Having overcome my initial resistance to querying my data through something other than SQL, Flux has grown on me and I get the point of it. I see how time series data needs a different language to talk to other than SQL. Thinking of data as streams is very useful, especially when you are faced with large quantities of it.
Ultimately I’d love to have the performance benefits of v3 while not sacrificing Flux. We have lots of data in your cloud offering and at times I am struggling with performance.
