30 6월 2016 출시

Beats 5.0.0-alpha4 released

By Monica Sarbu

Today we are excited to announce the fourth alpha release of the Beats 5.0.0 series.

IMPORTANT: This is an alpha release and is intended for testing purposes only. Please do not deploy in production.

Monitor MongoDB with Metricbeat

You are now able to monitor your MongoDB instance with Metricbeat. It gives you information about the uptime (ms) of the service, the number of active connections, the number of unused connections, the number of transactions written to the journal, the total size of the heap space (bytes), the number of page faults, the number of inserts, updates, deletes, the RAM size (MB), and many other metrics.

At the moment, Metricbeat has support for monitoring Apache, MySQL, Nginx, Redis, System statistics, Zookeeper, and MongoDB. If you are interested in monitoring another service, please open a feature request in GitHub or contribute to the open source community by adding a new Metricbeat module.

Support for gzip compression in Elasticsearch output

You can now configure the Beats to gzip encode the payload when using the Elasticsearch bulk API. The compression level is configurable in the Elasticsearch output section and must be in the range of 1 (best speed) to 9 (best compression). By default the gzip support is off (compression level is set to zero), so you need to explicitly opt in.

Ignore Symlinks log files

In the previous versions, Filebeat had support for symlinks to log files, but it often happened that the same log file was read twice. Now Filebeat is completely ignoring the symlinks

Remove topology_expire option

Prior to 5.0, you could configure the topology_expire option in Packetbeat to automatically expire the topology entries from Elasticsearch. Starting with Elasticsearch 5.0, the _ttl index option is removed, so the topology_expire option no longer works. We have removed the option in 5.0, and the user is responsible for any required cleanup.

Kibana Dashboards for Apache Metricbeat Module

Thanks to Radoondas, Metricbeat comes with a predefined Kibana Dashboard for the Apache Module.

Known issue and workaround

There is a known issue in the 5.0.0-alpha4 release of Kibana which affects the loading process of our sample dashboards. It only affects brand new installations of Kibana 5.0.0-alpha4 that don’t yet have any indices defined. If you are upgrading you can skip this. When following the Beats getting started guide, after loading the sample dashboards, you will get an error when trying to set the index pattern as the default:

Screen Shot 2016-06-30 at 10.19.52.png

If this happens, you can either manually overwrite the index pattern and then run the loading script again, or paste the following in the Kibana Console to force setting the default index (replacing `packetbeat-*` with the name of the Beat you are using):

POST .kibana/config/5.0.0-alpha4/_update
{
    "doc": {
        "defaultIndex": "packetbeat-*"
    }
}

Become a Pioneer

A big "Thank you!" to everyone who has tried the previous alpha releases and posted issues or provided feedback. We’d like to also remind you that if you post a valid, non-duplicate bug report during the alpha/beta period against any of the Elastic stack projects, you are entitled to a special gift package.