InfluxDB is a high performance time series database with efficient compression. The architecture is well designed (measurements retentions, shards) with easy ingestion, natively or through common time series protocols (OpenTSDB, Graphite…). Queries are SQL Like and the reporting is intuitive with Grafana or Chronograf. This introduction to InfluxDB v 1.7 also points out the breaks with the next major version 2.0.

The migration to InfluxDB v2 will need adjustments, many InfluxDB v1 functionalities are replaced in version 2. Migration procedure from version 1.x to version 2 is not yet unveiled, surely it will preconize export/import methods. The terminology is a little bit modified in version 2 : a database is now a bucket and an organization is necessarily attached to a bucket. No changes about the InfluxDB Line protocol. In this article : getting started with InfluxDB v2 and the breaks to manage when scheduling migration from version 1.x InfluQL is replaced by the language Flux, Continuous queries are replaced by tasks. Native support of the protocols Graphite, OpenTSDB… is removed, Telegraf will have to be used. Grafana plugin for Flux is not yet fully ready.

InfluxDB v2 has been released in November 2020. Upgrading from InfluxDB v1.8 to InfluxDB v2 is quite easy, bear in mind :

• A database + retention policy is a bucket in version 2.
• Backward compatibility with 1.x API is guaranteed, but only for authenticated 1.x users.
• Continuous queries must be migrated to Flux tasks.
• InfluQL is replaced by Flux language.
• Native support of OpenTSDB, Graphite… protocols is removed, Telegraf must be implemented.

An overview of the migration before being able to take full advantage of the Flux language features (joins, pivots, SQL packages…)

Moving from InfluxQL (SQL-Like) to InfluxDB v2 Flux language is confusing at the beginning, but you learn the mechanics very quickly. An overview of the migration of existing InfluxQL queries to Flux (filters, windowing, copying data, etc.).

InfluxDB v2 Flux language, quick reference guide and cheat sheet. Querying data, filters, windowing, joins, pivots, histograms, map, reduce, SQL data sources…

The new Flux language introduced in InfluxDB v2 addresses many InfluxQL language limitations. Overview of advanced Flux features : joins, pivots, histograms, computed columns (map), custom aggregate functions (reduce). In most topics, real use cases are discussed : joining data when timestamps differ by a few seconds, simulating outer joins waiting for the new join methods to come (left, right…), building histograms without cumulative data but the difference…

Great new feature of the InfluxDB v2 time series databases and its Flux language : the gateways to SQL databases engines (PostgreSQL, MySQL, Microsoft SQL Server…). From InfluxDB, to interact with SQL databases, 2 straightforward functions : sql.from and sql.to Use cases and specific considerations about data types conversions.