I suspect that I have damaged/corrupt NSS metadata on one of my volumes. It is a clustered resource running on OES2SP1 on robust hardware. I have 5 other volume resources exactly like it in terms of size, number of directories/files, amount of trustee assignments, etc.

The volume in question went down hard after a hardware failure and the pool had some difficulties coming back up. Eventually it did come back up. Now I am experiencing a variety of odd issues with it:

-metamig takes over 20 minutes to dump the trustees of this volume. It doesn't take as much as 5 seconds for any of the other similar volumes. It does however, dump all the rights.
-ncpcon nss resync also takes over 20 minutes for this volume. Once again it doesn't take as much as 5 seconds for any other volume
-Third-party utilities like JRB Utils also take 20+ minutes to read the trustees of the volume. This particular utility can't read half the trustees on the volume (it doesn't report them as existing) even though it works perfectly on all other volumes. Right-click "Trustee Rights...." does show the rights!
-Users have begun calling reporting that they don't have rights to certain directories any more. When I review them, rights are assigned correctly. However, I must remove the rights and re-add them to resolve these situations.
-The cluster resource now takes 15+ minutes to unload if it unloads successfully at all. It has begun going comatose with increasing frequency once these problems have been noticed. The problems exist no matter what physical node the resource is loaded on.

I have already tried a ravsui rebuild on the volume which has had no effect. I'm not sure what options I have left at this point. Any suggestions?