# Rotation of logs in production environments

**URL:** <https://discuss.pomerium.com/t/rotation-of-logs-in-production-environments/77>\
**Category:** Support\
**Created:** [February 16, 2022, 3:03am UTC](https://discuss.pomerium.com/t/rotation-of-logs-in-production-environments/77 "2022-02-16T03:03:11Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![asagayam](https://avatars.discourse-cdn.com/v4/letter/a/ba8739/32.png) [@asagayam](https://discuss.pomerium.com/u/asagayam)\
**Post date:** [February 16, 2022, 3:03am UTC](https://discuss.pomerium.com/t/rotation-of-logs-in-production-environments/77/1 "2022-02-16T03:03:11Z")

</div>

Just a general question. We are running pomerium in a production environment and we redirect the logs from stdout to a log file. We want to rotate logs since it’s growing quite large in a short span of time. We are considering using [multilog](http://cr.yp.to/daemontools/multilog.html) for this purpose. Are there any other alternatives to be considered?

---

<div class="post-metadata">

**Author:** ![calebdoxsey](https://avatars.discourse-cdn.com/v4/letter/c/7bcc69/32.png) [@calebdoxsey](https://discuss.pomerium.com/u/calebdoxsey)\
**Post date:** [February 16, 2022, 3:01pm UTC](https://discuss.pomerium.com/t/rotation-of-logs-in-production-environments/77/2 "2022-02-16T15:01:54Z")

</div>

Most linux distributions have switched to systemd for managing services. It includes log management via journalctl and can be customized: [journald.conf(5) — Arch manual pages](https://man.archlinux.org/man/journald.conf.5). It can also forward logs to syslog which has its own set of configuration options.

Another option is to use a log aggregator of some sort. For example google cloud has cloud logging: [https://cloud.google.com/logging/](https://cloud.google.com/logging/) or there are cross-platform 3rd party services like Datadog: [https://docs.datadoghq.com/logs/](https://docs.datadoghq.com/logs/).

You can also run your own log aggregator via Elasticsearch.