From 990a7a560c98dcbaa9c9e8deb0968819b646a664 Mon Sep 17 00:00:00 2001 From: David Timber Date: Mon, 16 May 2022 15:53:36 +0800 Subject: Changes ... - Deprecate palhm-dnssec-check.sh - Merge check-dnssec and boot-report config into the sample config - Add crontab sample - Reduce Python requirement to 3.5 - Remove use of capture_output - boot-report: remove systemd-analyze as the command is not available during boot time - Change config schema - "object-groups" and "objects" are now optional - Change "boot-report" include behaviour --- README.md | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'README.md') diff --git a/README.md b/README.md index 98e9361..1d87724 100644 --- a/README.md +++ b/README.md @@ -242,5 +242,15 @@ Also, you can always do a dry run of your backup task by setting the backend to ## TODO * JSON schema validation +### AWS S3 Replication Daemon +To prepare for very unlikely events of +[disasters](https://docs.aws.amazon.com/whitepapers/latest/disaster-recovery-workloads-on-aws/disaster-recovery-options-in-the-cloud.html) +affecting an entire AWS region, you may wish to implement cross-region +replication of S3 objects. The replication the S3 provides does not work on very +large objects. So replication of large objects across AWS regions has to be done +manually by a client - another implementation is required. + +Cross-region data transfer is costly, so this idea came to a halt. + ## Footnotes [^1]: Even with SSDs, disrupting sequential reads decreases overall performance -- cgit v1.2.3-70-g09d2