Elliswilson: Created page with "== Quick Reference == {| class="wikitable" |- ! Task !! Command |- | Check filesystem health || `lctl get_param health_check` |- | List filesystem space || `lfs df -h` |- | List MDT/OST space || `lfs df -i` (inodes), `lfs df -h` (blocks) |- | Check server status || `lctl dl` |- | Mount a client || `mount -t lustre mgsnode@tcp:/fsname /mnt/fsname` |- | Unmount a client || `umount /mnt/fsname` |-..."

2026-07-20T15:30:22Z

Created page with "== Quick Reference == {| class="wikitable" |- ! Task !! Command |- | Check filesystem health || <code>lctl get_param health_check</code> |- | List filesystem space || <code>lfs df -h</code> |- | List MDT/OST space || <code>lfs df -i</code> (inodes), <code>lfs df -h</code> (blocks) |- | Check server status || <code>lctl dl</code> |- | Mount a client || <code>mount -t lustre mgsnode@tcp:/fsname /mnt/fsname</code> |- | Unmount a client || <code>umount /mnt/fsname</code> |-..."

New page

== Quick Reference ==

{| class="wikitable"
|-
! Task !! Command
|-
| Check filesystem health || <code>lctl get_param health_check</code>
|-
| List filesystem space || <code>lfs df -h</code>
|-
| List MDT/OST space || <code>lfs df -i</code> (inodes), <code>lfs df -h</code> (blocks)
|-
| Check server status || <code>lctl dl</code>
|-
| Mount a client || <code>mount -t lustre mgsnode@tcp:/fsname /mnt/fsname</code>
|-
| Unmount a client || <code>umount /mnt/fsname</code>
|-
| Start a target || <code>mount -t lustre /dev/sdX /lustre/fsname/mdt0</code>
|-
| Stop a target || <code>umount /lustre/fsname/mdt0</code>
|}

== Startup Sequence ==

Always start services in this order:

<pre>
1. MGT (Management Target)
2. MDT (Metadata Targets — MDT0 first)
3. OSTs (Object Storage Targets)
4. Clients
</pre>

<pre>
# 1. Start MGS
mount -t lustre /dev/sda1 /lustre/mgt

# 2. Start MDT(s) — MDT0 must be first
mount -t lustre /dev/sdb1 /lustre/testfs/mdt0

# 3. Start OST(s)
mount -t lustre /dev/sdc1 /lustre/testfs/ost0
mount -t lustre /dev/sdd1 /lustre/testfs/ost1

# 4. Mount clients
mount -t lustre mgsnode@tcp:/testfs /mnt/testfs
</pre>

For ZFS backends, import the zpool before mounting:
<pre>
zpool import <poolname>
mount -t lustre <poolname>/mdt0 /lustre/testfs/mdt0
</pre>

'''Note:''' Clients connect to the MGS first, not the MDS. If the MDTs and OSTs are running but the MGS is down, clients cannot mount.

== Shutdown Sequence ==

Reverse of startup:

<pre>
1. Clients (umount all clients first)
2. MDT (stops new metadata operations)
3. OSTs
4. MGT (optional — can remain running for other filesystems)
</pre>

<pre>
# 1. Unmount clients
umount /mnt/testfs

# 2. Stop MDT(s)
umount /lustre/testfs/mdt0

# 3. Stop OST(s)
umount /lustre/testfs/ost0
umount /lustre/testfs/ost1

# 4. Stop MGS (optional)
umount /lustre/mgt
</pre>

For ZFS, it is ''not'' necessary to export zpools after unmounting. Only export if migrating the pool to another host.

== Mounting by Label ==

Lustre targets can be mounted by label instead of device path. Labels are set during <code>mkfs.lustre</code>:

<pre>
# Mount by label
mount -t lustre -L testfs-MDT0000 /lustre/testfs/mdt0
mount -t lustre -L testfs-OST0000 /lustre/testfs/ost0
</pre>

== fstab Entries ==

Server fstab entries should use <code>_netdev</code> to ensure networking is up before mount, and <code>noauto</code> to prevent mounting before LNet is configured:

<pre>
# Server targets
/dev/sdb1 /lustre/testfs/mdt0 lustre _netdev,noauto 0 0
/dev/sdc1 /lustre/testfs/ost0 lustre _netdev,noauto 0 0

# Client
mgsnode@tcp:/testfs /mnt/testfs lustre _netdev,noauto 0 0
</pre>

Alternatively, use labels:

<pre>
LABEL=testfs-MDT0000 /lustre/testfs/mdt0 lustre _netdev,noauto 0 0
</pre>

== Checking Filesystem Status ==

=== Health Check ===

<pre>
# On any server or client
lctl get_param health_check
</pre>

Returns <code>healthy</code> or lists active problems.

=== Filesystem Space ===

<pre>
# Block usage (human-readable)
lfs df -h /mnt/testfs

# Inode usage
lfs df -i /mnt/testfs
</pre>

=== Server Connectivity ===

<pre>
# List loaded Lustre devices
lctl dl

# Check LNET status
lctl get_param nis
lnetctl net show
</pre>

=== OST Status ===

<pre>
# Check if any OSTs are degraded or unavailable
lctl get_param obdfilter.*.degraded
lctl get_param osc.*.state
</pre>

== Common Mount Options ==

{| class="wikitable"
|-
! Option !! Applies To !! Description
|-
| <code>_netdev</code> || Servers, Clients || Wait for network before mounting.
|-
| <code>noauto</code> || Servers, Clients || Do not mount at boot (use with init scripts or systemd).
|-
| <code>flock</code> || Clients || Enable filesystem-wide POSIX flock (default: disabled).
|-
| <code>localflock</code> || Clients || Enable node-local flock only.
|-
| <code>user_xattr</code> || Clients || Enable user extended attributes.
|-
| <code>nouser_xattr</code> || Clients || Disable user extended attributes.
|-
| <code>lazystatfs</code> || Clients || <code>statfs</code> does not block if an OST is unavailable.
|}

== See Also ==

* [[Starting and Stopping Lustre Services]]
* [[Mounting a Lustre File System on Client Nodes]]
* [[Working with File System Labels]]
* [[Lustre Monitoring and Statistics Guide]]

[[Category:Lustre Systems Administration]]

Lustre Day to Day Operations - Revision history