[LU-6978] mkfs.lustre: LDISKFS-fs (loop0): Unrecognized mount option "force_over_128tb" or missing value

• Status: Resolved
• Project: Lustre
• Affects Version/s: Lustre 2.8.0
• Fix Version/s: Lustre 2.8.0
• Type: Bug
• Priority: Critical
• Reporter: Alexander Zarochentsev
• Assignee: Hongchao Zhang
• Resolution: Fixed
• Labels: patch

Environment

single node setup

• Severity: 3

Description

impossible to store force_over_128tb as a persistent option at mkfs.lustre time.

[root@devvm0 tests]# ../utils/mkfs.lustre --mgsnode=devvm0@tcp --reformat --fsname=foo --ost --mountfsoptions="user_xattr,errors=remount-ro,force_over_128tb" /dev/loop0
warning: /dev/loop0: for Lustre 2.4 and later, the target index must be specified with --index

   Permanent disk data:
Target:     foo:OSTffff
Index:      unassigned
Lustre FS:  foo
Mount type: ldiskfs
Flags:      0x72
              (OST needs_index first_time update )
Persistent mount opts: user_xattr,errors=remount-ro,force_over_128tb
Parameters: mgsnode=192.168.56.101@tcp

device size = 195MB
formatting backing filesystem ldiskfs on /dev/loop0
	target name  foo:OSTffff
	4k blocks     50000
	options        -I 256 -q -O extents,uninit_bg,dir_nlink,quota,huge_file,flex_bg -G 256 -E resize=4290772992,lazy_journal_init -F
mkfs_cmd = mke2fs -j -b 4096 -L foo:OSTffff  -I 256 -q -O extents,uninit_bg,dir_nlink,quota,huge_file,flex_bg -G 256 -E resize=4290772992,lazy_journal_init -F /dev/loop0 50000
mkfs.lustre: Unable to mount /dev/loop0: Invalid argument

mkfs.lustre FATAL: failed to write local files
mkfs.lustre: exiting with 22 (Invalid argument)
[root@devvm0 tests]# dmesg | tail
LDISKFS-fs (loop0): Unrecognized mount option "force_over_128tb" or missing value
[root@devvm0 tests]#

Issue Links

• is related to: LU-6662 mkfs should allow use of errors=panic
• is related to: LU-6030 cleanup for ldiskfs patches

Comments

[Alexander Zarochentsev]

After "LU-6030 osd-ldiskfs: improve mount option handling" landing, ldiskfs doesn't recognise
"mballoc","iopen","noiopen","iopen_nopriv","extents","noextents","bigendian_extents","force_over_128tb". Any attempt to store them as persistent options fails because mkfs.lustre mounts the fs image as ldiskfs with the same set of options it wants to store in mountdata file.

[Gerrit Updater]

Alexander Zarochentsev ([e-mail]) uploaded a new patch: http://review.whamcloud.com/15937
Subject: LU-6978 utils: mkfs.lustre to recognise non ldiskfs opts
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 5274b7522955b0193d3f98f8e7dcb21c0b53e4ca

[Alexander Zarochentsev]

proposed patch http://review.whamcloud.com/15937

[Andreas Dilger]

This is similar to LU-6662 and LU-6952 so I'm assigning it to Hongchao since he is also working on those issues.

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/15937/
Subject: LU-6978 utils: mkfs.lustre to recognise non ldiskfs opts
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 1d1e001d6cddc3890d8315636a7a8eb4fdecd932

[Peter Jones]

Landed for 2.8