nfs_refresh_inode: inode number mismatch (bug)

DISCUSSION Frank Cusack 1 participant 2 messages Jun 2, 2003

Subsystems (2)

NFS client NFS server

lore

Share & tools

lore

Frank Cusack

author

2003-06-02 08:37 UTC #0 lore raw

Hi,

I'm using a frankenstein kernel, 2.4.21-rc3 with some -ac bits, and 2.5.69 NFS+RPC backported to it. Like the CITI kernel (for krb5), but a little more aggressive on the bits backported.

I've found an issue with nfs_rmdir(), at least that's where I think it is. Consider these two shell sequences:

1 2

cd /nfs cd /nfs
mkdir tmp
echo foo > tmp/foo
less tmp/foo
[less waits for input]
rm -rf tmp
'v'
[vi tries to access tmp/foo]

At this point, inode.c:__nfs_refresh_inode() prints the "inode number mismatch" error. AFAICT, this is just noise, but the noise is driving me crazy. :-) I'm using fsstress and I can't tell if the messages are all because of this problem, or if my backport is bad as well.

Now, if sequence 2 is run on a different machine, there is no error! So that hints to me that the local cache just needs to be cleared, perhaps in nfs_rmdir() or maybe in nfs_unlink()/nfs_safe_remove(). I've tried a few things, but I'm not familiar enough with the code and am not making headway.

The stock 2.4.20 has this problem as well. I haven't tried any 2.5 kernels. The 2.2 kernel doesn't have this problem, because it doesn't allow you to unlink a .nfsXXX file while it's open (and therefore you cannot remove the dir); clearly incorrect behavior.

This is against a netapp server, although I can't see how the server would matter.

Thanks for any advice, guidance, or hopefully a fix! BTW, I'm interested to hear what tools folks use to stress NFS.

/fc

-------------------------------------------------------
This SF.net email is sponsored by: eBay
Get office equipment for less on eBay!
http://adfarm.mediaplex.com/ad/ck/711-11697-6916-5
_______________________________________________
NFS maillist - [email protected]
https://lists.sourceforge.net/lists/listinfo/nfs

Frank Cusack

author

2003-06-02 09:46 UTC #1 lore raw

Quoted 1 line · Frank Cusack

I've found an issue with nfs_rmdir(), at least that's where I think it is.

A little more info ...

This seems related to nfs3, and I'm going to guess readdirplus. If I mount as nfsv2, I don't see the problem.

It just struck me that there are a lot of readdirplus-related logs just before the "bug". I didn't think anything of it (because, *of course* there is a bunch of activity there) but in a rare moment of clarity I figured I should try v2.

I'll continue to look at this tomorrow, but I'm sure Trond or one of the wizards here will instantly know where the problem is. :-)

/fc