cloudalchemy / ansible-fluentd Goto Github PK

It would be nice to find a generic way of supporting fluentd plugins configuration via jinja templates.

Official Chef cookbook could probably help: https://github.com/treasure-data/chef-td-agent

Maybe current solution should be replaced by this one: https://github.com/fluent/fluent-plugin-prometheus

Maybe current solution should be replaced by this one: https://docs.fluentd.org/v1.0/articles/monitoring-prometheus
It looks like this one offers more functionality.

This role was initially created as an alternative to other roles that weren't installing plugin for prometheus metrics. Since fluentd has now native, configurable support for those metrics I think there is no need to compete with roles such as https://github.com/idealista/fluentd-role. It also seems like there is not much userbase and there is no active user of this role among maintainers. For those reasons, I think it is time to deprecate and archive this ansible role.

If there is no objection in 21 days (till 01.12.2019), I will archive the role with a following note in README:

We are no longer supporting this role, which means we will not accept and PRs nor new issues. We won't be removing this repository, but we strongly encourage you to switch to alternatives such as https://github.com/idealista/fluentd-role

/cc @jkrol2

Actually, the machine says:

$ ulimit -n
100001

Then, I changed file_descriptors.stdout < "1025" to file_descriptors.stdout < 1025, it works as expected

Somehow on centos td-agent service isn't enabled on boot.

Use ansible module instead of copying file for adding yum repository.

It would be great to add prometheus metrics exporter for fluentd. This exporter should be enabled with one variable fluentd_metrics: true.

Here is a link to the exporter: https://github.com/V3ckt0r/fluentd_exporter