[LU-471] mkfs.lustre ext4 extended options passed via --mkfsoptions are ignored

• Type: Bug
• Resolution: Fixed
• Priority: Minor
• Fix Version/s: Lustre 2.1.0, Lustre 1.8.7
• Affects Version/s: Lustre 2.1.0
• Environment: RHEL 5.6, e2fsprogs-1.41.90.wc2-0redhat, mkfs.lustre v2.0.62
• Severity: 3
• Assignee: Zhenyu Xu
• Reporter: John Hammond

Description

Using mkfs.ext4, I can set extended options using the -E flag:
mkfs.ext4 -b 4096 -E stride=64,stripe-width=384 -j -m 0 -J device=/dev/md25 /dev/md15

According to the manpage and mailing list http://comments.gmane.org/gmane.comp.file-systems.lustre.user/9766, I should be able to do the same for mkfs.lustre using the --mkfsoptions flag. However, mkfs.lustre calls mke2fs with its own set of extended options which causes mine to be ignored:

mke2fs -j -b 4096 -L punch-OST0000 -b 4096 -E stride=64,stripe-width=384 -j -m 0 -J device=/dev/md20 -I 256 -i 262144 -O extents,uninit_bg,dir_nlink,huge_file,flex_bg -G 256 -E resize=4290772992,lazy_journal_init, -F /dev/md10 732579456

The output of mkfs.lustre includes "Stride=0 blocks, Stripe width=0 blocks" suggesting that these options are ignored by mkfs.ext4.

Issue Links

• Trackbacks: Lustre 1.8.x known issues tracker While testing against Lustre b18 branch, we would hit known bugs which were already reported in Lustre Bugzilla https://bugzilla.lustre.org/. In order to move away from relying on Bugzilla, we would create a JIRA
• Trackbacks: Changelog 1.8 {}version 1.8.7wc1{} {}Support for networks: socklnd \any kernel supported by Lustre, qswlnd Qsnet kernel modules 5.20 and later, openiblnd IbGold 1.8.2, o2iblnd OFED 1.3, 1.4.1, 1.4.2, 1.5.1, 1.5.2, 1.5.3.1 and 1.5.3.2 gmlnd GM 2.1....