[LU-12731] mkfs.lustre Fails with message "invalid filesystem option set" when large_dir option is enabled

• Status: Closed
• Project: Lustre
• Type: Bug
• Priority: Minor
• Reporter: Arshad Hussain
• Assignee: WC Triage
• Resolution: Not a Bug

Environment

1. uname -a
   Linux devbox 3.10.0-862.9.1.el7_lustre.x86_64 #1 SMP Mon Aug 27 17:48:12 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux

Single Node lustre system compiled with commit aba68250a67a10104c534bd726f67b31a7f35692

MDT is rather small. Just used for dev purposes

1. ls -alih /tmp/lustre-mdt1
   2767607 rw-rr- 1 root root 245M Sep 8 00:46 /tmp/lustre-mdt1

• Severity: 3

Description

Using llmount.sh, which calls mkfs.lustre fails on MDT with invalid filesystem option. Simply removing the "large_dir" option make it work. Since we have the workaround making it minor.

$ /root/mrp/lustre-xx-3.10/lustre-release/lustre/tests/../utils/mkfs.lustre --mgs --fsname=lustre --mdt --index=0 --param=sys.timeout=20 --param=mdt.identity_upcall=/root/mrp/lustre-xx-3.10/lustre-release/lustre/tests/../utils/l_getidentity --backfstype=ldiskfs --device-size=250000 --mkfsoptions="-O ea_inode,large_dir" --reformat /tmp/lustre-mdt1

Permanent disk data:
Target: lustre:MDT0000
Index: 0
Lustre FS: lustre
Mount type: ldiskfs
Flags: 0x65
(MDT MGS first_time update )
Persistent mount opts: user_xattr,errors=remount-ro
Parameters: sys.timeout=20 mdt.identity_upcall=/root/mrp/lustre-xx-3.10/lustre-release/lustre/tests/../utils/l_getidentity

formatting backing filesystem ldiskfs on /dev/loop0
target name lustre:MDT0000
kilobytes 250000
options -I 1024 -i 2560 -q -O ea_inode,large_dir,dirdata,uninit_bg,^extents,dir_nlink,quota,huge_file,flex_bg -E lazy_journal_init -F
mkfs_cmd = mke2fs -j -b 4096 -L lustre:MDT0000 -I 1024 -i 2560 -q -O ea_inode,large_dir,dirdata,uninit_bg,^extents,dir_nlink,quota,huge_file,flex_bg -E lazy_journal_init -F /dev/loop0 250000k
Invalid filesystem option set: ea_inode,large_dir,dirdata,uninit_bg,^extents,dir_nlink,quota,huge_file,flex_bg

mkfs.lustre FATAL: Unable to build fs /dev/loop0 (256)

mkfs.lustre FATAL: mkfs failed 256

 

Good Case (without large_dir)

$ /root/mrp/lustre-xx-3.10/lustre-release/lustre/tests/../utils/mkfs.lustre --mgs --fsname=lustre --mdt --index=0 --param=sys.timeout=20 --param=mdt.identity_upcall=/root/mrp/lustre-xx-3.10/lustre-release/lustre/tests/../utils/l_getidentity --backfstype=ldiskfs --device-size=250000 --mkfsoptions="-O ea_inode" --reformat /tmp/lustre-mdt1

Permanent disk data:
Target: lustre:MDT0000
Index: 0
Lustre FS: lustre
Mount type: ldiskfs
Flags: 0x65
(MDT MGS first_time update )
Persistent mount opts: user_xattr,errors=remount-ro
Parameters: sys.timeout=20 mdt.identity_upcall=/root/mrp/lustre-xx-3.10/lustre-release/lustre/tests/../utils/l_getidentity

formatting backing filesystem ldiskfs on /dev/loop0
target name lustre:MDT0000
kilobytes 250000
options -I 1024 -i 2560 -q -O ea_inode,dirdata,uninit_bg,^extents,dir_nlink,quota,huge_file,flex_bg -E lazy_journal_init -F
mkfs_cmd = mke2fs -j -b 4096 -L lustre:MDT0000 -I 1024 -i 2560 -q -O ea_inode,dirdata,uninit_bg,^extents,dir_nlink,quota,huge_file,flex_bg -E lazy_journal_init -F /dev/loop0 250000k
Writing CONFIGS/mountdata

 

Comments

[Andreas Dilger]

What version of e2fsprogs do you have installed?

[Arshad Hussain]

Thanks Andreas!
I was using old (ancient) version of 1.42.13. Updating to latest 1.45.2 solves the error. Closing the bug.

[Arshad Hussain]

"large_dir" options did not work with e2fsprogs 1.42.13. Updating e2fsprogs to latest 1.45.2 solves the problem.