What a blast from the past this is! Thats my error message, so sorry for it not being helpful.
We assumed it wouldn't get hit my many people and that it would be quite a while before it was, looks like that time has come :-) Did increasing the size of the dupreq cache help avoid the problem ? Out of interest, how many clients, filesystems are involved here? This probably was initially found on a system with about 15TB of data spread out over about 1500 file systems, it was a 10CPU E10k that just did NFS (assuming I remember it correctly). This message posted from opensolaris.org