iac/ansible/roles/s3fs/defaults/main.yaml
Sergei Poljanski 66ab6d8db3 s3fs: allow turning the mounts off, off on asxpio-ops
On the 38 GB ops disk the s3fs cache reserve (ensure_diskfree) never
fit, so all three mounts failed at every boot since 2026-09-22. Backups
now go to Hetzner S3 with rclone instead.

Add s3fs_enabled (default true); when false the role unmounts the
buckets and drops their fstab entries, keeping the credentials file
that the rclone wrapper reads.

Assisted-by: Claude:opus-5
2026-10-12 05:03:11 +04:00

56 lines
1.4 KiB
YAML

---
# S3FS role defaults
# S3 Configuration
s3_endpoint: "https://{{ s3_region }}.your-objectstorage.com"
s3_region: "hel1"
# S3 Credentials (should be overridden in group_vars or vault)
s3_access_key: ""
s3_secret_key: ""
# false: no mounts. Unmounts and drops the fstab entries; the credentials
# file stays (backups use it with rclone, /root/backups/rclone-hetzner.sh).
s3fs_enabled: true
# S3 Buckets to mount (non-replication buckets only)
s3_buckets:
- name: asxpio-terraform-main
mount_point: /mnt/s3/terraform
owner: root
group: root
mode: "0755"
- name: asxpio-backups-main
mount_point: /mnt/s3/backups
owner: root
group: root
mode: "0755"
- name: asxpio-immich-main
mount_point: /mnt/s3/immich
owner: root
group: docker
mode: "0775"
# S3FS mount options
# Note: s3fs_ensure_diskfree is defined per-host in host_vars/
s3fs_options:
- "passwd_file=/etc/s3fs/.passwd-s3fs"
- "use_path_request_style"
- "url={{ s3_endpoint }}"
- "allow_other"
- "umask=0022"
- "uid={{ s3fs_uid | default('0') }}"
- "gid={{ s3fs_gid | default('0') }}"
- "use_cache=/var/cache/s3fs"
- "ensure_diskfree={{ s3fs_ensure_diskfree | default(81920) }}"
- "max_stat_cache_size=10000"
- "stat_cache_expire=900"
- "parallel_count=15"
- "multipart_size=52"
- "nocopyapi"
- "enable_noobj_cache"
# Package dependencies
s3fs_packages:
- s3fs
- fuse3