Posted on February 10, 2009 at 10:23 am

And Linux in General, F-U.

During shutdown

Inode 0000010029fa7bc8: orphan list check failed!

Well crap.  Guess we’ll be fscking on bootup….this is going to take a while….

/dev/VolGroup00/vz: Inode 80987262, i_blocks is 163148, should be 162952.  FIXED.
/dev/VolGroup00/vz: Inode 99082779, i_blocks is 11422, should be 11420.  FIXED.
/dev/VolGroup00/vz: Inode 115564804, i_blocks is 740918, should be 740916.  FIXED.
/dev/VolGroup00/vz: Inode 136136891 has illegal block(s).  

/dev/VolGroup00/vz: UNEXPECTED INCONSISTENCY; RUN fsck MANUALLY.
	(i.e., without -a or -p options)
[FAILED]

WELL GEE YA THINK!?  IS THAT NOT YOUR WHOLE POINT OF EXISTENCE FSCK? TO FIX UNEXPECTED INCONSISTENCIES?!  I’m sorry people but seriously, this little thing about Linux fsck has ALWAYS bugged the crap out of me.  Especially since it will only ever come up after running fsck Pass 1 (which usually takes a while, and on a big set of RAID drives like this, over an HOUR).  OK fine I’ll fsck you manually….

(Repair filesystem) 1 # fsck -y /dev/VolGroup00/vz
fsck 1.35 (28-Feb-2004)
e2fsck 1.35 (28-Feb-2004)
/dev/VolGroup00/vz contains a file system with errors, check forced.
Pass 1: Checking inodes, blocks, and sizes
Inode 136136891 has illegal block(s).  Clear? yes

Illegal block #27404 (4180300720) in inode 136136891.  CLEARED.
Illegal block #27405 (2533802954) in inode 136136891.  CLEARED.
………
(and another hour or so later)
Too many illegal blocks in inode 891297853.
Clear inode? yes

Inode 900579329 was part of the orphaned inode list.  FIXED.
Restarting e2fsck from the beginning...
/dev/VolGroup00/vz contains a file system with errors, check forced.
Pass 1: Checking inodes, blocks, and sizes
…….
THE HELL? So now it’s on it’s *THIRD* Pass 1.  I’m absolutely CERTAIN that with only a LITTLE programming effort by someone who knows the filesystem that that restart there that just cost yet more HOURS was completely unnecessary.  FURTHER since the POINT of fsck (especially when the filesystem was MARKED DIRTY) is to check for and FIX ANY UNEXPECTED CONSISTENCY ERRORS it should’ve just DONE it in the first damned place.

This frustration is exactly why ZFS is going to hopefully wipe Linux out.  Even the Solaris UFS tools aren’t THIS damned bad.  And I’ve put a ZFS system through far more hell than this machine will ever see, and came out with less downtime.