mkfs (hfs): Warning - What?!

[Lisa Mauer]

Never seen this warning before - anyone have any info for me?

Putting in everything I did, warning is at the bottom
*****************************************
# lvcreate vg02
Logical volume "/dev/vg02/lvol3" has been successfully created with
character device "/dev/vg02/rlvol3".
Volume Group configuration for /dev/vg02 has been saved in /etc/lvmconf/vg02.con
f
# lvextend -L 1024 /dev/vg02/lvol3 /dev/dsk/c5t1d0
# pvdisplay /dev/dsk/c5t1d0
--- Physical volumes ---
PV Name /dev/dsk/c5t1d0
VG Name /dev/vg02
PV Status available
Allocatable yes
VGDA 2
Cur LV 1
PE Size (Mbytes) 4
Total PE 4340
Free PE 2048
Allocated PE 2292
Stale PE 0
IO Timeout (Seconds) default
Autoswitch On

# lvextend -L 1024 /dev/vg02/lvol3 /dev/dsk/c5t1d0
Logical volume "/dev/vg02/lvol3" has been successfully extended.
Volume Group configuration for /dev/vg02 has been saved in /etc/lvmconf/vg02.con
f
# newfs -F hfs -m 1 /dev/vg02/rlvol3
mkfs (hfs): Warning - inode blocks/cyl group (74) >= data blocks (32) in last
mkfs (hfs): cylinder group. This implies 256 sector(s) cannot be allocated.
mkfs (hfs): /dev/vg02/rlvol3 - 1048320 sectors in 1680 cylinders of 16 tracks, 39 sectors
1073.5Mb in 105 cyl groups (16 c/g, 10.22Mb/g, 1600 i/g)
Super block backups (for fsck -b) at:
16, 10040, 20064, 30088, 40112, 50136, 60160, 70184, 80208, 90232,
100256, 110280, 120304, 130328, 140352, 150376, 159760, 169784, 179808, 189832,
199856, 209880, 219904, 229928, 239952, 249976, 260000, 270024, 280048, 290072,
300096, 310120, 319504, 329528, 339552, 349576, 359600, 369624, 379648, 389672,
399696, 409720, 419744, 429768, 439792, 449816, 459840, 469864, 479248, 489272,
499296, 509320, 519344, 529368, 539392, 549416, 559440, 569464, 579488, 589512,
599536, 609560, 619584, 629608, 638992, 649016, 659040, 669064, 679088, 689112,
699136, 709160, 719184, 729208, 739232, 749256, 759280, 769304, 779328, 789352,
798736, 808760, 818784, 828808, 838832, 848856, 858880, 868904, 878928, 888952,
898976, 909000, 919024, 929048, 939072, 949096, 958480, 968504, 978528, 988552,
998576, 1008600, 1018624, 1028648, 1038672
#bdf -l
/dev/vg02/lvol3 1025617 9 1015351 0% /opt/dir

[Helen French]

Hi,

I see a problem here:

1) If you are going to mirror the LV, then you have to use 'lvextend -m 1' instead of 'newfs -m 1'.

what actually are you doing ?

HTH,
Shiju

[Sridhar Bhaskarla]

Looks like you didn't quite mention the commands you ran. mkfs with a -m 1 option should return error and do nothing.

-Sri

[Lisa Mauer]

The -m is for free_space_percent,
The minimum percentage of free disk space allowed. The default value is 10 percent, we only want 1 percent. Not for mirroring.

[Lisa Mauer]

I ran
# newfs -F hfs -m 1 /dev/vg02/rlvol3

[Jeff Schussele]

Appears to me that the "-m 1" parameter on the newfs command is incorrect.
Mirroring is at the LV level not the FS.
I'm just wondering why it didn't kick out an error msg. Undocumented option?
Now the -m on mkfs does display the command line that was used to create an existing FS & I believe that's what you're seeing here - but again the 1 should have kicked an error & it wants the dev not the LV.
Also I thought newfs/mkfs work on the /dev/dsk/cxtydz device & not the LV.

Rgds,
Jeff

[Helen French]

Hi,

Check the man pages. The -m option in mkfs is used for: - Display the command line that was used to create the file system. The file system must already exist. This option provides a means to determine the parameters used to construct the file system.

I think you have to use option '-o' with 'minfree' value, for the specific output you needed.

# man 1m mkfs_hfs

HTH,
Shiju

[Jeff Schussele]

If you want 1% free you would use either:

-o minfree,1 OR
-o minfree=1

I'm not sure as I don't normally use HFS.

See the following docs on mkfs (hfs):

http://docs.hp.com/cgi-bin/fsearch/framedisplay?top=/hpux/onlinedocs/B2355-90681/B2355-90681_top.html&con=/hpux/onlinedocs/B2355-90681/00/01/191-con.html&toc=/hpux/onlinedocs/B2355-90681/00/01/191-toc.html&searchterms=mkfs&queryid=20020409-124304

I still think it interpreted the -m wrong & that caused the problem.

HTH,
Jef

[MANOJ SRIVASTAVA]

Hi L.Mauer


do a newfs -F vxfs /dev/vg02/rlvol3 before doing onay lvextend operations , ie incase you want to reintialize liek you are doing it .

- m1 option is used for mirroing and is to be used with lvextend command and not with newfs command.


Manoj Srivastava

[Tom Danzig]

There's nothing wrong with his newfs command. From the newfs_hfs man page:

-m free_space_percent
The minimum percentage of free disk space allowed. The default value is 10 percent. Once the file system capacity reaches this
threshold, only users with appropriate privileges can allocate disk blocks.

I do not, however, know what the warning mean. It appears that the file system was created.

[Jeff Schussele]

Well it appears that L. Mauer & Tom are correct here & the man pgs really suck in this situation.
ONLY the newfs_hfs man pages show this NOT the newfs pgs
Looking at the msg I now believe it's just warning that the last cylinder group is small & cannot be fully used - essentially saying that there are 256 sectors less than normally usable. I believe Tom is correct & the FS WAS created.

See this is why I'm a VXFS person ;~)

Rgds,
Jeff

[Sachin Patel · Accepted solution]

Hi Lisa,
We uses hfs and also -m options but like

newfs -F hfs -b 65536 -f 8192 -o largefiles -t 32 -c 32 /dev/vg22/rhp3

then
tunefs -e 624 -m 1 /dev/vg22/rhp1

Now you will have 1% reserve rather then 10%.

Sachin Patel

[Michael Tully]

Hi,

I understand the answer, but I'm wondering why you would want to create a HFS file system in the first place. The JFS or vxfs filesystem is more more efficient and you don't have to fool around with the min free options.

Cheers
~Michael~

PS: "You can tune a file system, but you can't tune a fish"