[LU-4057] sub-directory NFS reexport issue

• Status: Resolved
• Project: Lustre
• Affects Version/s: Lustre 2.5.0
• Type: Bug
• Priority: Major
• Reporter: Dmitry Eremin (Inactive)
• Assignee: Dmitry Eremin (Inactive)
• Resolution: Duplicate

Environment

RHEL 6.4, Lustre master

• Severity: 3

Description

If on server reexport sub-directory instead of root it not visible through NFS. There is no error during mount from both side NFS and Lustre but mount point is not valid.

cd /net/l201u2/mnt/lustre && ls export
d????????? ? ?    ?    ?            ? export

stat /net/l201u2/mnt/lustre/export
stat: cannot stat `/net/l201u2/mnt/lustre/export': Stale file handle

Issue Links

• is related to: LU-2904 parallel-scale-nfsv3: FAIL: setup nfs...
• is related to: LU-3550 Stale file handle on mount when mount...

Comments

[John Hammond]

I can't get past the cd. Did you use no_subtree_check in /etc/exports? Can you post the /etc/exports and mount options you used?

[Dmitry Eremin (Inactive)]

On Lustre node I launch llmount.sh and then service nfs start.

# mount
l201u2@tcp:/lustre on /mnt/lustre type lustre (rw,user_xattr,flock)
# cat /etc/exports
/mnt/lustre/export 192.168.1.0/24(rw,no_root_squash,no_all_squash)

[Dmitry Eremin (Inactive)]

On NFS node:

# mount
l201u2:/mnt/lustre/export on /net/l201u2/mnt/lustre/export type nfs (rw,nosuid,nodev,intr,sloppy,vers=4,addr=192.168.1.2,clientaddr=192.168.1.1)
# ls -la /net/l201u2/mnt/lustre
ls: cannot access /net/l201u2/mnt/lustre/export: Stale file handle
total 0
dr-xr-xr-x 3 root root 0 Oct  5 16:02 .
dr-xr-xr-x 3 root root 0 Oct  5 16:02 ..
d????????? ? ?    ?    ?            ? export

[Dmitry Eremin (Inactive)]

The no_subtree_check option don't change behavior.

# cat /etc/exports
/mnt/lustre/export 192.168.1.0/24(rw,no_root_squash,no_all_squash,no_subtree_check)

[Dmitry Eremin (Inactive)]

The root cause is described in LU-2904:
nfs defects, it does not work if the "inode::i_ino" is larger than 2^32, which is known issue.