Thomas Graichen wrote:
> "Nathan Scott" <nathans@xxxxxxxxxxxxxxxxxxxxxxxx> wrote:
> > hi Thomas,
>
> > On Nov 29, 9:21am, Thomas Graichen wrote:
> >> Subject: Re: stress test on ppc
> >> "Nathan Scott" <nathans@xxxxxxxxxxxxxxxxxxxxxxxx> wrote:
> >>
> >> ...looks like it happens pretty early
> >>
> >> ppc:/usr/src/xfs/cmd/xfs/mkfs # ./mkfs.xfs -f /dev/hda9
> >> meta-data=/dev/hda9 isize=256 agcount=8, agsize=49152
> >> blks
> >> data = bsize=4096 blocks=393216, imaxpct=25
> >> = sunit=0 swidth=0 blks, unwritten=0
> >> naming =version 2 bsize=4096
> >> log =internal log bsize=4096 blocks=1200
> >> realtime =none extsz=65536 blocks=0, rtextents=0
> >> ppc:/usr/src/xfs/cmd/xfs/mkfs # xfs_db -r -c sb -c p /dev/hda9|grep
> >> blocksize
> >> blocksize = 524288
> >> ppc:/usr/src/xfs/cmd/xfs/mkfs #
> >>
>
> > ok, so the problem is in either mkfs or xfs_db. after mkfs, run:
>
> > # od -c -N 8 /dev/hda9
> > 0000000 X F S B \0 \0 020 \0
> > 0000012
>
> > you should see exactly that (the second 4 are the blocksize).
> > if you do, its an xfs_db problem. if not, its a mkfs problem.
>
> ppc:~ # od -c -N 8 /dev/hda9
> 0000000 X F S B \0 \0 020 \0
> 0000010
> ppc:~ #
>
> so it looks like xfs_db
>
> > lemme know. (at the same time, could you send the unfiltered
> > "xfs_db -r -c sb -c p /dev/hda9" output also?)
>
> no problem
Ok this doesn't look good.
Several fields looked bad already.
Notably the rootino should be 128
Was this output taken directly after a mkfs or after a mount?
We need to narrow down our problems.
First we need to figure out if mkfs is correctly
making the file system.
Second is xfs_db correctly reporting the values
on disk.
Finally where in XFS are we trashing AG blocks.
Ok lets do things in this order
mkfs a file system
od -c -N 8 -j 512 /dev/sdb1
should result in
0001000 X A G F \0 \0 \0 001
od -c -N 8 -j 1024 /dev/sdb1
and
od -c -N 8 -j 1024 /dev/sdb1
0002000 X A G I \0 \0 \0 001
Those are the magic #s
now try
xfs_db
sb 0
print
agf 0
print
agi 0
print
save output
Mount the file system
run od commands and xfs_db command again
save output
Now do some FS activity
run od command and xfs_sb commands again
save output.
Send all the output to us.
This should tell us where we need to start fixing things first.
I suspect we will need to find an alpha around someplace to actually
get some of this stuff debugged.
>
>
> ppc:~ # xfs_db -r -c sb -c p /dev/hda9
> magicnum = 0x58465342
> blocksize = 524288
> dblocks = 105553116266496
> rblocks = 0
> rextents = 0
> uuid = bc3a44f0-947a-4ebe-964e-5d237161432b
> logstart = 262148
> rootino = 72057594037927936
> rbmino = 129
> rsumino = 4683743612465315840
> rextsize = 16
> agblocks = 196608
> agcount = 8
> rbmblocks = 0
> logblocks = 1200
> versionnum = 0x2084
> sectsize = 64
> inodesize = 256
> inopblock = 2048
> fname = "\000\000\000\000\000\000\000\000\000\000\000\000"
> blocklog = 12
> sectlog = 144
> inodelog = 8
> inopblog = 32
> agblklog = 16
> rextslog = 0
> inprogress = 0
> imax_pct = 152
> icount = 64
> ifree = 13546827679130451968
> fdblocks = 391980
> frextents = 0
> uquotino = 0
> pquotino = 0
> qflags = 0
> flags = 0
> shared_vn = 0
> inoalignmt = 1073741824
> unit = 0
> width = 0
> dirblklog = 0
> ppc:~ #
>
> t
>
> --
> thomas.graichen@xxxxxxxxxxxxxx
> technical director innominate AG
> clustering & security the linux architects
> tel: +49-30-308806-13 fax: -77 http://www.innominate.com
--
Russell Cattelan
cattelan@xxxxxxxxxxx
|