Author: Dana Medhaug
Updated: January 2025
Audience: Everyone
Environmental details: Self-hosted Replicated - Traditional/Native application server preparation
Summary
In the past, when preparing an application server for Native Scheduler Replicated, we recommended that customers start with a 100GB - 200GB drive and partition the drive into Logical Volumes.
Process
1. Start with a new 200 GB disk (unpartitioned)
2. Create mount points:
mkdir /data /logs /var/lib/docker /var/lib/replicated
3. Create Physical Volume:
pvcreate /dev/sdb
(sdb is an example here; this would be the name of your new disk)
4. Create Volume Group:
vgcreate vg_jama /dev/sdb
5. Create Logical Volumes (example for 200 GB disk, increase if a larger disk is used, leaves 10GB for growth):
lvcreate -L 40G -n lv_docker vg_jama
lvcreate -L 30G -n lv_replicated vg_jama
lvcreate -L 20G -n lv_logs vg_jama
lvcreate -L 100G -n lv_data vg_jama
6. Write File systems:
mkfs.xfs -L docker -n ftype=1 /dev/vg_jama/lv_docker
mkfs.ext4 -L replicated /dev/vg_jama/lv_replicated
mkfs.ext4 -L data /dev/vg_jama/lv_data
mkfs.ext4 -L logs /dev/vg_jama/lv_logs
7. Edit /etc/fstab:
LABEL=docker /var/lib/docker xfs defaults 0 0
LABEL=replicated /var/lib/replicated ext4 defaults 0 0
LABEL=data /data ext4 defaults 0 0
LABEL=logs /logs ext4 defaults 0 0
8. Mount all volumes:
mount -a
Feedback: Please leave feedback in the comments below.
Related to
Comments
0 comments
Please sign in to leave a comment.