Return to BSD News archive
Path: sserve!manuel!munnari.oz.au!network.ucsd.edu!dog.ee.lbl.gov!ucbvax!bloom-beacon!bloom-picayune.mit.edu!news.mit.edu!raeburn From: raeburn@cambridge.cygnus.com (Ken Raeburn) Newsgroups: comp.unix.bsd Subject: 386bsd nfs hang problem -- some data, temporary workaround Message-ID: <RAEBURN.92Jul21181551@cambridge.cygnus.com> Date: 21 Jul 92 22:15:59 GMT Sender: news@athena.mit.edu (News system) Organization: Cygnus Support, Cambridge MA Lines: 18 Nntp-Posting-Host: cambridge.cygnus.com I reported previously that I was seeing problems reading files over NFS using the ne2000 driver; timeouts would eventually be reported, no data would be read. Listing files and directories (small ones anyways) were not a problem. After playing with etherfind and kernel printfs, I've come to this conclusion: Fragmented 8K UDP packets from the NFS server are not reaching the UDP layer in 386bsd. The Sun is sending them (according to another Sun spying on the network), but the UDP input routine is never called. I don't know if the bug here is on the 386bsd or Sun side, and won't have time to look into it in the next couple of days. In the meantime, mounting NFS file systems with "rsize=1024" does get rid of this problem. (It does nothing about TCP being slow, though.) Ken