# V2.3 and Output pipeline
**URL:** https://discuss.elastic.co/t/v2-3-and-output-pipeline/48689
**Category:** Logstash
**Created:** [April 28, 2016, 2:29pm UTC](https://discuss.elastic.co/t/v2-3-and-output-pipeline/48689 "2016-04-28T14:29:29Z")
**Posts on this page:** 4
**Page:** 1
### Author:  [@phr0gz](https://discuss.elastic.co/u/phr0gz)
#### Post date: [April 28, 2016, 2:29pm UTC](https://discuss.elastic.co/t/v2-3-and-output-pipeline/48689/1 "2016-04-28T14:29:29Z")
Hello, is it possible to modify an output and reload the config without losing
all events in main memory ?
e.g.
In case of having this configuration:
Input: tcp
Output: elasticsearch, file, tcp
Problem: Disk full, The "file output blocking" will block elasticsearch and tcp outputs ( if any output is down, all output will stall).
One solution will be to clean the disk!
But is it possible to change the file "path" to another disk and reload (not a "force-stop") the config, to avoid data loss?
---
### Author:  [@magnusbaeck](https://discuss.elastic.co/u/magnusbaeck)
#### Post date: [April 30, 2016, 3:16pm UTC](https://discuss.elastic.co/t/v2-3-and-output-pipeline/48689/2 "2016-04-30T15:16:33Z")
This currently isn't possible. Persistent queues will be added in the context of [https://github.com/elastic/logstash/issues/2605](https://github.com/elastic/logstash/issues/2605) so it should work once that's in.
---
### Author:  [@phr0gz](https://discuss.elastic.co/u/phr0gz)
#### Post date: [April 30, 2016, 3:41pm UTC](https://discuss.elastic.co/t/v2-3-and-output-pipeline/48689/3 "2016-04-30T15:41:49Z")
Great, thx for the feedback
---
### Author:  [@system](https://discuss.elastic.co/u/system)
#### Post date: [July 6, 2017, 4:59am UTC](https://discuss.elastic.co/t/v2-3-and-output-pipeline/48689/4 "2017-07-06T04:59:52Z")