Grafana v6 has been released - new features of an open source visualization tool.
At the end of February, the sixth version of the Grafana monitoring platform was released. In the material we will tell more about the features of this release and the new features of the tool.
/ Flickr / Simon / CC BY-SA
Grafana is a tool for data visualization. It can work with tables, graphs, text, maps and supports integration with third-party applications and plug-ins. Grafana makes it possible to create so-called dashboards with panels, each of which displays the specified system indicators for a set period. Each dashboard is universal and can be customized to the needs of a specific project.
')
Such companies as NetApp, PayPal Stack Overflow and Red Hat work with Grafana. In particular, PayPal with Grafana monitors the performance of services, and the Red Hat team helps the platform to monitor the load on the OpenStack Cloud.
The repository of the project can be found on GitHub , and on the official site - documentation and quick start guide .
New version of Grafana presented at the conference GrafanaCon, which was held in late February in Los Angeles. You can read about all the innovations of Grafana v6 in the change log on the official project website. We will tell about the most interesting of them.
Switch from Angular to React. The community requested to provide an opportunity to write custom plug-ins on React in the spring of last year. In the new release, the developers of Grafana have implemented this feature. Now the @ grafana / ui SDK library will be used for plugins.
At the same time, Grafana made a complete migration to the new library. According to the developers, they took this step to simplify the support of the code base (to make the code future-proof).
Added Explore . This is a tool for analysis and debugging that will allow you to find out the causes of infrastructure problems. It integrates with Prometheus Query Editor. Usually, the user contacts the Prometheus server using the PromQL query language to obtain information about the status of a particular object. Grafana v6 will allow you to bypass this step and immediately display all the information on dashboards.
Also, in the updated editor, the function of auto-filling of the names of metrics and queries has earned, and filters have appeared for searching data in tables. According to the developers, these updates will help you quickly find the information you need without switching between tools.
New logging system Loki . This is a horizontally scalable logging system, which does not index the contents of the logs, but their metadata. Users were able to store logs in compressed format. This approach will be useful when working with Kubernetes pod logs, since their tags are indexed automatically.
Removed user session storage. This was done to make it easier to work with software like Redis, Memcache, Postgres, or MySQL. Instead of sessions , tokens of authentication will now use tokens with a short life cycle. Tokens life time users can configure in the configuration file. The developers have also implemented several security updates: cookies, by default, work with the SameSite attribute, which protects users from the Cross-Site Request Forgery ( CSRF ) attack.
Among the less significant improvements is also worth noting the change in the design of the color palette and support for Google Hangouts chat notifications.
/ Flickr / Bruce Guenter / CC BY
Grafana is not the only open source data visualization tool. As an alternative solution, you can select the Kibana platform, which is included in the ELK stack (ElasticSearch, Logstash, Kibana). That is what we use in 1cloud as a logging tool - we described our ELK implementation experience in a separate article .
Among the advantages of the Kibana community is the ease of deployment and the ability to text search. Grafana is also appreciated for a nice and functional interface. In this case, Kibana, unlike Grafana, is able to analyze text messages logs. For this reason, Kibana is more suitable if the user is faced with the task of examining the system information and determining with its help the cause of the fault.
In addition, Kibana wins over Grafana in terms of scalability. Last year, the Punchplatform development team abandoned Grafana in favor of Kibana when the number of virtual servers increased to several hundred.
The disadvantages of Kibana include limited integration. This tool only works with Elasticsearch, and Grafana can be used with third-party services like Prometheus or Redis.
According to BI-SURVEY forecasts , solutions for the collection and visual presentation of information will be one of the main trends in 2019. It is considered that the industry is waiting for the transition from simple data display systems in graph and table format to “smart” tools based on AI systems. This approach will automate a number of routine tasks associated with monitoring IT infrastructure.
For this reason, it can be expected that Grafana, Kibana and similar projects will become more flexible and acquire components that will not only monitor, but also independently resolve incidents.
/ Flickr / Simon / CC BY-SA
What is Grafana
Grafana is a tool for data visualization. It can work with tables, graphs, text, maps and supports integration with third-party applications and plug-ins. Grafana makes it possible to create so-called dashboards with panels, each of which displays the specified system indicators for a set period. Each dashboard is universal and can be customized to the needs of a specific project.
')
Such companies as NetApp, PayPal Stack Overflow and Red Hat work with Grafana. In particular, PayPal with Grafana monitors the performance of services, and the Red Hat team helps the platform to monitor the load on the OpenStack Cloud.
The repository of the project can be found on GitHub , and on the official site - documentation and quick start guide .
What has changed in the new version
New version of Grafana presented at the conference GrafanaCon, which was held in late February in Los Angeles. You can read about all the innovations of Grafana v6 in the change log on the official project website. We will tell about the most interesting of them.
Switch from Angular to React. The community requested to provide an opportunity to write custom plug-ins on React in the spring of last year. In the new release, the developers of Grafana have implemented this feature. Now the @ grafana / ui SDK library will be used for plugins.
At the same time, Grafana made a complete migration to the new library. According to the developers, they took this step to simplify the support of the code base (to make the code future-proof).
Added Explore . This is a tool for analysis and debugging that will allow you to find out the causes of infrastructure problems. It integrates with Prometheus Query Editor. Usually, the user contacts the Prometheus server using the PromQL query language to obtain information about the status of a particular object. Grafana v6 will allow you to bypass this step and immediately display all the information on dashboards.
Also, in the updated editor, the function of auto-filling of the names of metrics and queries has earned, and filters have appeared for searching data in tables. According to the developers, these updates will help you quickly find the information you need without switching between tools.
New logging system Loki . This is a horizontally scalable logging system, which does not index the contents of the logs, but their metadata. Users were able to store logs in compressed format. This approach will be useful when working with Kubernetes pod logs, since their tags are indexed automatically.
Removed user session storage. This was done to make it easier to work with software like Redis, Memcache, Postgres, or MySQL. Instead of sessions , tokens of authentication will now use tokens with a short life cycle. Tokens life time users can configure in the configuration file. The developers have also implemented several security updates: cookies, by default, work with the SameSite attribute, which protects users from the Cross-Site Request Forgery ( CSRF ) attack.
Among the less significant improvements is also worth noting the change in the design of the color palette and support for Google Hangouts chat notifications.
/ Flickr / Bruce Guenter / CC BY
Grafana Analogs
Grafana is not the only open source data visualization tool. As an alternative solution, you can select the Kibana platform, which is included in the ELK stack (ElasticSearch, Logstash, Kibana). That is what we use in 1cloud as a logging tool - we described our ELK implementation experience in a separate article .
Among the advantages of the Kibana community is the ease of deployment and the ability to text search. Grafana is also appreciated for a nice and functional interface. In this case, Kibana, unlike Grafana, is able to analyze text messages logs. For this reason, Kibana is more suitable if the user is faced with the task of examining the system information and determining with its help the cause of the fault.
In addition, Kibana wins over Grafana in terms of scalability. Last year, the Punchplatform development team abandoned Grafana in favor of Kibana when the number of virtual servers increased to several hundred.
The disadvantages of Kibana include limited integration. This tool only works with Elasticsearch, and Grafana can be used with third-party services like Prometheus or Redis.
“Grafana and Kibana are the tools that will help solve different tasks,” said Sergey Belkin, head of the development department at IaaS provider 1cloud.ru . - If we are talking about monitoring several hundred or thousands of servers, then Kibana will do better. If the primary task is integration with third-party services, then Grafana is worth paying attention to. ”
The future of data visualization solutions
According to BI-SURVEY forecasts , solutions for the collection and visual presentation of information will be one of the main trends in 2019. It is considered that the industry is waiting for the transition from simple data display systems in graph and table format to “smart” tools based on AI systems. This approach will automate a number of routine tasks associated with monitoring IT infrastructure.
For this reason, it can be expected that Grafana, Kibana and similar projects will become more flexible and acquire components that will not only monitor, but also independently resolve incidents.
Some materials from our corporate blog:
Source: https://habr.com/ru/post/443006/
All Articles