May 2018 Summaries

DevOps is a complex and multifaceted concept that can be challenging to define due to its varying approaches and requirements across different organizations, particularly between large and small companies. Despite these challenges, DevOps aims to bridge the gap between development and operations teams by promoting collaboration, automation, and continuous improvement. By extracting insights from technology and effective communication, organizations can successfully implement DevOps strategies that meet their unique needs and resources. Ultimately, building a strong company culture is crucial for achieving long-term success in DevOps, requiring adaptability and flexibility in strategy implementation.

The InfluxDays London event will be held on June 14 at The Brewery in London, focusing on the impact of time series data and covering topics such as metrics and event data from sensors, containers, systems, and microservices. Attendees will come from a wide variety of roles including CTOs, developers, architects, and engineers from diverse industries, with presentations and real-world experiences from innovative companies like Amazon Web Services, CJC, and CERN. The event aims to provide educational and informative content for attendees, packed with opportunities for spirited discussions and professional networking, and is part of InfluxData's expansion into the European market. InfluxData has built its developer and customer base across industries by delivering a modern Open Source Platform for analyzing metrics and events data, enabling customers to derive better business insights and take on digital transformation initiatives.

InfluxQL and Flux have unified schema exploration concepts, allowing users to explore a schema using tags keys and values. InfluxQL meta queries can be used to show measurement names, list all the tag keys for a specific measurement, list distinct values for a specific tag in a measurement, list fields for a measurement, and expand on these results to see available values or keys within those expansions. Users can use helper functions to simplify their queries and Flux plans to formalize additional metadata querying capabilities in the near future.

Enabling The Things Network for InfluxDB`, an article by David G. Simmons, outlines a simple process for connecting The Things Network to InfluxDB, highlighting its ease of integration and the brevity of the resulting post, which was attributed to the simplicity of the connection.

InfluxData has been selected to present at several industry events in June, including dotScale, Open Source Data Center Conference, and IoT Slam Live 2018, to address growing interest in its time series platform knowledge and expertise. The company will be presenting on topics such as building out InfluxDB Cloud 2.0, distributed monitoring, and pushing IoT analytics to the edge. InfluxData is recognized as the leader among Time Series Database management systems, enabling customers to quickly build monitoring, IoT applications, and real-time analytics applications. With over 400 customers across industries, including manufacturing, financial services, energy, and telecommunications, InfluxData is pioneering the shift to time series data in a modern metrics and events platform.

InfluxDB 1.5.3 and InfluxDB Enterprise 1.5.3 maintenance releases are now available, addressing important defect fixes recommended for immediate application by InfluxData to ensure stability and security. The open source release binaries can be found on the downloads page, while enterprise customers should log in to their portal to download the updated binaries. Additionally, users previously using TSI preview with earlier versions of InfluxDB Enterprise need to follow specific upgrade steps to take advantage of this feature in production use.

The hinted handoff (HH) queue is a crucial component of eventual consistency in distributed systems like InfluxDB Enterprise, ensuring that data replication factor for each database is achieved by storing and replicating writes to the HH queue when a node goes offline. The queue acts as a safe place for unsuccessful writes to land, allowing nodes to complete writes once they come back online, thereby achieving eventual consistency. Understanding the HH queue is essential for troubleshooting and identifying problematic patterns in distributed systems, which will be discussed in the next article.

Kubernetes, an open-source container orchestration system, has raised concerns about its complexity and whether it will collapse under its own weight. The author questions whether Kubernetes is truly connected with the needs of its users, particularly application developers who are not yet benefiting from its scalability and self-healing capabilities. Most applications do not require such complexity, and the majority of developers work on modest scale projects or those seeking product-market fit, which can lead to code being thrown away if it doesn't meet user demand. The author suggests that Kubernetes' cognitive load in early project stages is too high, making it difficult for developers to focus on feature velocity and iteration speed. To overcome this, the author proposes introducing generators and scaffolding tools similar to those used in Rails, which enabled massive developer productivity gains. A streamlined experience would allow developers to ship features faster and make Kubernetes a successful development platform. The CNCF project landscape is also becoming increasingly complex, making it challenging for application developers to navigate and choose the right tools for their projects. Ultimately, the author believes that infrastructure should optimize application developer efficiency and ability to deliver real user problems, and that a well-designed open platform will win over others.

Chronograf 1.5.0 and Kapacitor 1.5.0 have been released, featuring enhanced data visualization capabilities in Chronograf, including the ability to visualize tabular data using a table graph type, and improved alert sending functionality for both platforms, with support for multiple Slack workspaces and updated APIs for OpsGenie and PagerDuty. Additionally, Kapacitor has introduced new alert inhibitors that allow suppressing events from other matching alerts, while also utilizing React 16 under the hood for improved performance in Chronograf.

The TICK Stack's Chronograf has been updated to version 1.5, which introduces a modern platform for metrics and events, enabling users to visualize both metrics and events within the same interface. This release also includes tabular views of time series data, allowing users to view their text-based event data alongside metrics in a single dashboard. Kapacitor 1.5 has been released simultaneously, adding support for Kafka and Slack workspaces, as well as various minor improvements based on customer feedback. The updates are designed to provide a more holistic view of metrics and events, making it easier for users to analyze their data and integrate with downstream systems.

Telegraf 1.6.3 has been released, marking a maintenance update for the popular monitoring system. The new version includes several plugin updates, including fixes for intermittent panics in Aerospike Input and a potential wait group deadlock in Apache Input. Additionally, the Librato Output plugin has been updated to support unsigned integer and boolean data types. This release is available for download on the Telegraf downloads page.

This article by Noah Crowley from DZone provides a step-by-step guide on how to install and configure the TICK Stack, which consists of Telegraf, InfluxDB, Chronograf, and Kapacitor, on Windows. Although Windows support is considered experimental for all components except Telegraf, the full TICK Stack can still be installed and run on the platform. The article includes a detailed walkthrough of each component's installation on Windows, making it easier for users to set up the stack on this operating system.

Boston Time Series Meetup at Wayfair` The Boston Time Series Meetup at Wayfair was a successful event with around 45 attendees despite severe thunderstorms in the area. The meetup featured talks by Jim Hagan from Wayfair, who discussed their environment and data pipeline using Graphite and InfluxDB, and Ryan Betts from InfluxData, who provided an update on InfluxDB Internals and the unique characteristics of Time Series Databases. Wayfair's use of both platforms to collect, deliver, and analyze metrics from over 2,000 VMs running hundreds of applications was highlighted as a key aspect of their operations. The event also included discussions on next-generation pipelines using Kafka and Telegraf streaming service, and upcoming talks by Ben Bianchi of Wayfair and Jacob Lisi from Grafana.

Steve Moreton, Senior Technical Director at CJC, will be speaking at InfluxDays London 2018 about mosaicOA, a big data visualization platform used exclusively in the capital markets industry to support real-time data for infrastructure. The platform was developed to handle large data volumes and provides the necessary granularity to manage mission-critical infrastructure. CJC chose InfluxDB as the metrics store behind mosaicOA, enabling it to provide insights that were previously unobtainable. Steve will present a live demo of mosaicOA at InfluxDays London on June 14, 2018, where attendees can learn more about this innovative solution.

The TICK Stack is a set of open-source tools for monitoring and managing applications, and it can be run on Windows as an experimental platform. The stack consists of InfluxDB, Kapacitor, Chronograf, and Telegraf. To install the TICK Stack on Windows, developers can extract the archives to specific directories, such as C:\Program Files\Chronograf, C:\Program Files\InfluxDB, and so on. Each application has its own executable file that can be run using a Command Prompt. The default configurations are sufficient for most users, but customizations can be made by modifying configuration files or environment variables. The applications expose specific ports at startup, such as HTTP API endpoints, web GUIs, and more. Data is stored on disk in subdirectories of the user's home directory. Telegraf supports running as a Windows service, while InfluxDB does not currently support this feature. Alternative approaches for running the TICK Stack include downloading the source code from GitHub, using Docker, Virtual Machines, or the Windows Subsystem for Linux. Developers can reach out to the community site or Twitter @InfluxDB for specific questions or support.

Enabling The Things Network for InfluxDB` InfluxDB can be easily integrated with The Things Network, a low-power wide-area network (LPWAN) platform used for IoT applications. To do this, a Telegraf instance is installed on a server and configured to read from the TTN MQTT broker by editing the telegraf.conf file. This configuration allows for easy data streaming from TTN sensors into InfluxDB, where it can be analyzed and visualized using various tools and dashboards. The integration process is straightforward and requires minimal setup, making it an attractive option for users of TTN looking to analyze their data in a more structured format. With the Telegraf MQTT plugin, the JSON payload from the TTN MQTT broker is automatically flattened and imported into InfluxDB with no additional effort required, streamlining the analysis process.

The Large Hadron Collider (LHC) at CERN is a powerful tool used to study the fundamental structure of the universe by smashing particles together at high speeds. The LHC generates massive amounts of data, which is collected and analyzed using various sensors and applications. To ensure efficient operation and prevent overheating, the LHC relies on monitoring systems that track temperature levels, energy use, and structural health. These systems are supported by databases, networks, and querying tools, requiring careful maintenance to address any failures or disruptions caused by external factors such as rodents or birds. CERN's efforts to monitor and manage its critical projects have been highlighted through the use of InfluxDB, a database that has played a crucial role in optimizing the efficiency of the LHC's 2000 nodes processing data at an incredible rate.

InfluxData has stepped up its speaking engagements, with a focus on time series data monitoring and analysis, as the interest in this topic continues to grow.

InfluxDB Enterprise clusters consist of two types of nodes: meta nodes, which manage state consistency, and data nodes, where raw time series data is stored. To achieve high availability, a cluster requires an odd number of meta nodes, with at least three being recommended for optimal performance. Data nodes are used to store raw data, while replication factor determines the number of copies of any one shard that should exist. The hardware requirements for meta nodes include 1-2 CPU cores, 512MB-1GB RAM, and a single HDD, whereas data nodes require more powerful hardware with at least 8+ CPU cores, 64+ GB RAM, and SSD drives (>1000 IOPS recommended). Load balancers can be used to distribute incoming requests across the cluster.

Telegraf 1.6.2 has been released, featuring updates to several plugins including CrateDB Output, Datadog Output, and JSON Parser, as well as fixes for issues such as ignoring the UTF8 BOM character in JSON parsing and a doc example issue with name_suffix vs name_override. The release also includes support for unsigned integers and improvements to the System Input plugin. This maintenance release is now available on the downloads page.

In 2018, CRN published a list of the 45 coolest data management and integration vendors, with InfluxData ranked among them. The company's CEO, Evan Kaplan, was profiled by Rick Whiting, who highlighted InfluxData's extensive stack of open-source technologies designed to manage time-series data from IoT networks and other systems. The InfluxData platform offers tools and services for IoT, DevOps monitoring, and real-time analytics use cases, addressing the challenges of managing continuous flow of time-series data.

InfluxData's CTO Paul Dix has been selected to present at KubeCon + CloudNativeCon Europe 2018, a conference being held in Copenhagen, Denmark. He will lead the session "Integrating Prometheus and InfluxDB" on May 4, where he will describe how InfluxDB can serve as a long-term store of record for Prometheus data, thanks to recent additions adding support for Prometheus standards and protocols. The presentation aims to help attendees integrate these two systems effectively.

InfluxData is expanding its speaking engagements to address growing interest in time series platform knowledge and expertise, particularly around IoT and DevOps monitoring and analytics. The company's modern Open Source Platform is designed for metrics, events, and other time-based data, enabling developers to build next-generation monitoring, analytics, and IoT applications. With over 400 customers across various industries, including manufacturing, financial services, energy, and telecommunications, InfluxData is pioneering the shift to time series in a modern metrics and events platform, making it possible for customers to become data-driven and take on digital transformation initiatives.

In the current "measure everything" era, CIOs must navigate the vast amounts of data generated by their companies' measurement systems to find actionable insights that can drive business success. To achieve this, successful CIOs need to focus on revenue-producing activities, provide real insights rather than just reports, and create systems that allow for timely action. They also need to invest in flexible systems that can grow and evolve alongside the company, and enable their organization to become a data-driven business.

InfluxData has announced that its Founder and CTO Paul Dix will be presenting at KubeCon + CloudNativeCon Europe 2018, where he will discuss the latest updates to Prometheus support, including a new store and record functionality for InfluxDB. Mr. Dix's session will focus on how InfluxDB can integrate with Prometheus, enabling users to perform federated queries against multiple servers. Additionally, InfluxData will unveil new functionalities around Kubernetes and provide an early view of its new query language IFQL. The company has rapidly built its developer and customer base across various industries by delivering a modern Open Source Platform that enables customers to derive better business insights from their metrics and events data.

This is a patch release of Chronograf, addressing an issue with CSV template variables being overwritten in certain instances. The new version is available for download from the project's downloads page.