home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.os.linux
- Path: sparky!uunet!pipex!warwick!sunserver1.aston.ac.uk!uhura!evansmp
- From: evansmp@uhura.aston.ac.uk (Mark Evans)
- Subject: Re: NFS query
- Message-ID: <1993Jan11.121047.1305@aston.ac.uk>
- Sender: usenet@aston.ac.uk (Usenet administrator)
- Nntp-Posting-Host: uhura
- Organization: Aston University
- X-Newsreader: TIN [version 1.1 PL241235]
- References: <1993Jan11.000948.9562@csd.uwe.ac.uk>
- Date: Mon, 11 Jan 1993 12:10:47 GMT
- Lines: 34
-
- Operator (Phil/Dylan) (root@slave.uwe.ac.uk) wrote:
- : I should really be posting this (not followup) but i can't 'cos
- : my smtp server (csd) won't recognize me (hummm) anyway
- : I have a query about NFS. Has anyone else experienced a problem
- : when transfering files to an NFS mount, in that occasionally
- : it will throw up an error about the NFS file handle being stale
- : in which case it will forget about the file entirely leaving
- : only a 0 length file entry on the NFS mount. This appears to
- : happen with cp rcp ftp tftp so I think the problem is burried
- : in the NFS code
-
- If you can, then login to the NFS server machine to copy files using
- ftp and such like.
- That way you will will avoid sending packets to your machine, then sending
- the same data straight out again, at least doubling your traffic.
- :
- : comments? fixes?
- It is possible that you may be generating lots of collisions, try bringing
- up a network analyser package (MAC level preferably) to see what is
- going on.
- (right that is possible for rcp, ftp, tftp. Except cp, the only time this
- sort of thing could be caused by cp is when both source and desination are
- on NFS partitions.)
- Maybe altering the timeouts in /etc/fstab will help.
-
- :
- : Phil
- : =--=
-
- --
- -------------------------------------------------------------------------
- Mark Evans |evansmp@uhura.aston.ac.uk
- +(44) 21 429 9199 (Home) |evansmp@cs.aston.ac.uk
- +(44) 21 359 6531 x4039 (Office) |
-