# Filebeat output resilience/scalability **URL:** \ **Category:** Beats\ **Tags:** filebeat\ **Created:** [May 29, 2020, 9:49am UTC](https://discuss.elastic.co/t/filebeat-output-resilience-scalability/234893 "2020-05-29T09:49:51Z")\ **Posts on this page:** 2\ **Page:** 1
**Author:** ![corentin.savineau](https://avatars.discourse-cdn.com/v4/letter/c/5f9b8f/32.png) [@corentin.savineau](https://discuss.elastic.co/u/corentin.savineau)\ **Post date:** [May 29, 2020, 9:49am UTC](https://discuss.elastic.co/t/filebeat-output-resilience-scalability/234893/1 "2020-05-29T09:49:51Z")
Hi there, I've been working on a cluster Elastic stack cluster. Everything works fine but I would like to push the resilience/scalability further. Regarding filebeat, I understood that I can put a "list" of ES nodes in order to make my filebeat client always sending logs whathever happens with cluster nodes. But what if all nodes of this list go offline for some reason? I certainly wouldn't be able to change the elastic output section on ALL my filebeat clients to put other ES nodes. So what are the options offered to me? Can I use proxy/loadbalancer as elastic output in my filebeat.yml? Or should I better use some kind of centralized management tool such as Ansible or Puppet in order to change that hosts: [IP,IP,IP,IP] in the ouput.elasticsearch section on all my servers at once? I am doing some R&D with Elastic Stack 7.7 but I am thinking about all the aspect of putting this to production with a lot of servers to monitor (about 500). If someone has got any advices I would be so grateful ---
**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\ **Post date:** [June 26, 2020, 9:49am UTC](https://discuss.elastic.co/t/filebeat-output-resilience-scalability/234893/2 "2020-06-26T09:49:52Z")
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.