27 messages in org.freebsd.freebsd-hackersNFS client/buffer cache deadlock| From | Sent On | Attachments |
|---|---|---|
| Brian Fundakowski Feldman | Apr 14, 2005 10:07 pm | |
| Marc Olzheim | Apr 15, 2005 6:21 am | |
| Brian Fundakowski Feldman | Apr 15, 2005 8:21 am | |
| Marc Olzheim | Apr 18, 2005 2:25 am | |
| Marc Olzheim | Apr 19, 2005 6:32 am | |
| Brian Fundakowski Feldman | Apr 19, 2005 8:18 am | |
| Marc Olzheim | Apr 19, 2005 9:02 am | |
| Marc Olzheim | Apr 19, 2005 9:09 am | |
| Brian Fundakowski Feldman | Apr 19, 2005 9:17 am | |
| Brian Fundakowski Feldman | Apr 19, 2005 1:48 pm | |
| Marc Olzheim | Apr 20, 2005 7:04 am | |
| Brian Fundakowski Feldman | Apr 20, 2005 7:26 am | |
| Marc Olzheim | Apr 20, 2005 7:39 am | |
| Brian Fundakowski Feldman | Apr 20, 2005 8:22 am | |
| Marc Olzheim | Apr 20, 2005 8:35 am | |
| Brian Fundakowski Feldman | Apr 20, 2005 8:54 am | |
| Jilles Tjoelker | Apr 20, 2005 10:12 am | |
| Brian Fundakowski Feldman | Apr 20, 2005 10:31 am | |
| Brian Fundakowski Feldman | Apr 20, 2005 11:03 am | |
| Marc Olzheim | Apr 20, 2005 11:03 am | |
| Dag-Erling Smørgrav | Apr 21, 2005 1:36 am | |
| Garrett Wollman | Apr 21, 2005 4:50 am | |
| Garrett Wollman | Apr 21, 2005 4:51 am | |
| Garrett Wollman | Apr 22, 2005 5:49 am | |
| Brian Fundakowski Feldman | Apr 22, 2005 8:12 am | |
| Brian Fundakowski Feldman | Apr 22, 2005 8:38 am | |
| Garrett Wollman | Apr 23, 2005 5:09 am |
| Subject: | NFS client/buffer cache deadlock | |
|---|---|---|
| From: | Marc Olzheim (marc...@stack.nl) | |
| Date: | Apr 20, 2005 8:35:10 am | |
| List: | org.freebsd.freebsd-hackers | |
On Wed, Apr 20, 2005 at 11:20:38AM -0400, Brian Fundakowski Feldman wrote:
Reads should be totally unaffected...
The server was misbehaving. Fixed. :-)
Btw.: I'm not sure write(),writev() and pwrite() are allowed to do short writes on regular files... ?
Our manpage is incorrect; POSIX states that they are (see earlier e-mail). There really is no alternative -- we simply can't build an NFS transaction larger than our buffer cache can accomodate. Note that short wries won't happen for normal buffer sizes, only excessively large ones. I really don't believe that writev() is meant to be used so that you can write gigantic data structures in a single transaction...
Ah, I was reading the SUSv2 page:
http://www.opengroup.org/onlinepubs/009695399/functions/write.html
instead of the POSIX version.
But in neither of those I can extrude the fact that it can return with result < nbyte, without it being a permanent condition. What phrase makes you conclude that it can ?
Marc