March 2017 Summaries

In this week's recap, InfluxDB and TICK-stack related issues, workarounds, how-tos, and Q&A from GitHub, IRC, and the InfluxData Community were discussed. Advanced conditionals in InfluxQL allowed for calculating minimum values across distinct time intervals while suppressing results where the minimum value is zero. Non-SELECT queries in Chronograf enabled user management without relying on the deprecated web admin console or HTTP API. Queries across measurements in InfluxDB allowed for querying field keys that appear only in some of the measurements, with the query returning no results for those missing measurements.

I joined InfluxData to lead the team building InfluxDB, after spending 8 years as CTO at VoltDB, a high-velocity ACID relational database developer. I was drawn to InfluxData's mission to make high-performance data simple for developers, addressing an unsolved problem in real-time metrics and IoT data management. The company prioritizes open source foundations, sustainable business practices, and talent-driven development teams.

The InfluxData stack, which includes tick-charts, has undergone significant updates with the goal of making it easier to run on Kubernetes. The primary change is the re-architecture of the telegraf chart, which has been split into two for improved usability and reduced complexity. This allows users to easily spin up telegraf instances while minimizing configuration requirements. The new telegraf-s chart will utilize Helm's `toToml` template function to simplify instance creation, making it easier to generate configurations for plugins like SNMP or Jolokia. Future improvements aim to enhance production readiness, user experience, and provide more examples, including monitoring Prometheus endpoints with Telegraf.

The latest beta update of Chronograf brings several exciting features to its users, including the creation of customized dashboards with a new `DASHBOARDS` page, improved cell creation and editing capabilities, and enhanced database management through the introduction of the `DB Management` tab on the `ADMIN` page. The update also includes better viewing options for dashboards, such as auto-refresh intervals, time ranges, and full-screen mode. Additionally, Chronograf continues to evolve with new features like customizable graph types and a more intuitive user interface.

InfluxDB has recently encountered several unexpected issues that require attention from users, including authentication failures with the Command Line Interface (CLI), data truncation in graphs when using Chronograf, and changes in query behavior. A recommended solution for authentication failures is to single-quote passwords if they contain special characters, as some shells may attempt to inject variables into the password. The default configuration option `max-row-limit` in OSS version 1.2.1 sets a limit of 10,000 rows returned per query, but setting it to 0 allows for unlimited rows and is recommended. Additionally, there has been a change in query behavior that returns fewer points than expected, with a workaround available using InfluxQL's subqueries until the issue is fixed in version 1.3.0.

The Case of The Enigmatic Enumerable Module` is a story about a developer named Dawn who was trying to update a Ruby gem called Facetwit in her application, but encountered an unexpected issue that caused her application's dashboard to display an error message. After investigating the problem, she discovered that the `each` method from Ruby's Enumerable module was being used in a way that caused multiple external API requests to be made without her knowledge or control, resulting in a "Too Many Requests" error. To solve the issue, Dawn realized that the `take` method from the same module could be used to limit the number of results fetched, and she successfully implemented this change to resolve the problem. The story highlights the importance of understanding how Ruby's Enumerable module works and using it responsibly when writing external API gems.

InfluxDB Cloud has been upgraded to use the latest InfluxDB Enterprise version 1.2.2, which benefits from the open source community's extensive testing and exploration of the underlying TICK stack components. The upgrade leverages customer deployments and operational situations within InfluxDB Cloud to further harden the release and deliver a higher quality product for both on-premise and cloud customers. The "digital exhaust" generated by customers using InfluxDB Cloud provides valuable insights for the engineering team, enabling them to focus efforts on improving stability, reliability, and overall experience. The upgrade process has been designed to minimize friction and de-risk new features, with a focus on simplifying the upgrade process and addressing operational challenges before they manifest in customer environments.

In this post, the authors from InfluxData share tips for using InfluxDB effectively, including recommendations for writing historical data, setting timestamp precision for writes, and finding log files in various operating systems. For sparse historical data, users can lengthen their retention policy's shard group duration and temporarily lower the cache-snapshot-write-cold-duration configuration setting to make the process more efficient. The authors also explain how to specify a second-precision timestamp when writing data using InfluxDB's CLI, and provide instructions for accessing log files on different operating systems, including systemd and System V.

InfluxData has announced the immediate availability of its cloud-based real-time monitoring and analytics platform, InfluxDB Cloud 1.2, which includes performance and visualization enhancements such as a 50% increase in write performance, integration with Grafana for more beautiful metric and analytic dashboards, enhanced query language support for sub-queries, improved data cache, and advanced alert condition handling. These upgrades are designed to improve scalability, availability, and security while empowering developers to build next-generation monitoring, analytics, and IoT applications faster, easier, and to scale, delivering real business value quickly.

InfluxDB Week in Review March 13, 2017` The InfluxDB team has released version 1.2 with improved subqueries and write throughput capabilities, resulting in a 50% increase on larger hardware configurations. This update is accompanied by various community-driven content, including blog posts, videos, and tutorials, as well as announcements about new hires and the launch of Influx University. Additionally, Fincity, a financial industry pioneer, is testing InfluxDB's TICK stack for monitoring server resources in AWS. The company has been using InfluxDB to power its platform since 2016. The community can also claim an InfluxDB hoodie as a reward for featuring their own applications or products on the official website and social media channels.

Helm is a templating packaging and deployment manager for Kubernetes that simplifies the management of Kubernetes applications by providing a simple, easy-to-use place to define, install, and upgrade complex applications. Helm charts are collections of files that describe related groups of Kubernetes resources, including metadata, templates, and values files. To create a Helm chart, one can use the `helm create` command, which generates a chart directory with common files and directories. The process of creating and updating Helm charts is straightforward, allowing developers to easily manage their applications and deploy them in any environment. With Helm, developers can create complex applications that include servers, caches, and more, making it an essential tool for Kubernetes users.

In this weekly recap, the most interesting InfluxDB and TICK-stack related issues, workarounds, and how-tos from GitHub, IRC, and the InfluxData Community are summarized. The nuances of microsecond syntax in InfluxDB were discussed, including a known issue where the proper microsecond syntax differs for writes, queries, the CLI's precision command, and the configuration file. Additionally, the syntax for sub-sub-queries was explained, highlighting that queries support several nested subqueries but not multiple SELECT statements per subquery. Furthermore, the importance of double-quoting identifiers in queries was emphasized, particularly when measurement names include special characters or start with digits. The TICK-stack's downloads are now live, and InfluxDB Cloud and Enterprise features are highlighted as well as a call to share testimonials from companies using InfluxDB.

InfluxDB has recently released version 1.2, which includes subqueries and improved write throughput on larger hardware configurations. The company is experiencing an increasing demand for monitoring solutions, and its technology is being used by various organizations such as Voiture Expert to monitor machinery inputs like temperature and humidity. InfluxDB is also expanding its community support with new features and resources, including live virtual training sessions and a plugin ecosystem. Additionally, the company has announced upcoming events and opportunities for users to share their success stories and get featured on its website.

InfluxData has created a new community forum called community.influxdata.com where users can ask and answer questions about InfluxDB usage, code samples, and general topics related to the stack. This move aims to provide a central platform for users to discuss technical issues, share ideas, and collaborate with each other. The forum will be moderated by InfluxData's technical team, but it is intended to be a community-driven space where users can engage with each other and receive support from peers.

In this article, the authors describe how to set up Kapacitor and Chronograf to send alert messages to Telegram using an API access token and chat ID. To start, users need to create a Telegram bot by following specific steps, including creating a username, description, and profile picture for the bot. They also need to obtain an API access token from the bot and identify their unique chat ID within the Telegram application. Next, they set up Kapacitor's configuration file with the `enabled` setting set to true and the `token` and `chat-id` settings populated with the obtained values. In a TICKscript, users include the `telegram()` attribute and specify the chat ID to send alert messages to Telegram. Finally, Chronograf provides an easy-to-use interface for creating alert rules and sending messages to event handlers like Telegram. The process involves selecting the Telegram option in the Kapacitor configuration page and entering the API access token and chat ID.