Inode number inconsistencies on UFS1 filesystem

[Christopher Nehren]

I'm running 5.1-R, and having forgot to set up UFS2 for my file systems 
(doh!) I'm using ACLs the UFS1 way. All is well and good, for the most 
part, except that I'm getting a lot of messages like these:


/var/log/messages:Sep 22 20:38:01 <kern.crit> prophecy kernel: 
ufs_extattr_get (/var): inode number inconsistency (-1493839883, 
-1493839880)
/var/log/messages:Sep 22 20:38:01 <kern.crit> prophecy kernel: 
ufs_extattr_get (/var): inode number inconsistency (-462848829, -462848826)
/var/log/messages:Sep 22 20:38:21 <kern.crit> prophecy kernel: 
ufs_extattr_rm (/var): inode number inconsistency (-462848829, -462848826)

These messages _seem_ to show up when the system is under heavy load, 
but I haven't done extensive testing to verify that. What's peculiar, 
though, is that I haven't actually set up any extended attributes on my 
/var partition as of yet, but I have on /usr/home and it's not showing 
these errors. I checked the source, and from what I can see, it appears 
that the extended attribute checking code is expecting to find extended 
attribute(s) on the inodes specified -- but because no attributes exist 
on the entire partition, that check of course fails, with ENOATTR. The 
data on the partition appears to be fine: I've fscked it multiple times, 
and haven't seen any errors. If there's anything else that I can do to 
check the file system's integrity, I'd be delighted to learn of it and 
try whatever it is. Thanks in advance for any assistance with this.