NFS Stale File Handles
Imad Succari
Has anyone ever played around with NFS and ways to distinguish
NFS Stale File Handles when they occur. I am trying to write
a program that will enable our users to log even if they are
trying to access something that has gone stale and therefore
cause them to hang. However, none of the files in /etc/ seem
to reflect a change of status on the filesystem even though it
has gone stale. If anyone has any experience in this area or
has any advive they can lend me I would greatly appreciate it.
Thank you in advance
Imad R. Succari
Programmer/Analyst
Dana-Farber Cancer Institute
Division of Biostatistics
Boston, MA. 02115
Thomas Sippel - Dau
In article <2rr4m4...@early-bird.think.com>, isuc...@ttest.dfci.harvard.edu (Imad Succari) writes:
- Has anyone ever played around with NFS and ways to distinguish
- NFS Stale File Handles when they occur.
Not a lot you can do, stale file handles are inherent in the design of NFS.
The problem is that all the file management is done on the NFS requestor,
the responder (or server) does as its told.
Thus, if you (on machine fred) mount a file system from a server bill,
and then log into a further client joe, change to a certain directory
on joe, and then recursively remove (from fred) a parent directory of
what you are looking at from joe, then the next time you try to do
some file access (or ls ...) from joe, you get "stale file handle".
(Thats one way of getting it, there are others.)
If you tried to do the same on just one or two machines, the software
would probably look through its inode caches and stop you from removing
active (i.e. current) directories. This behaviour is useful in that a
crash on a client will not cause dangling links on the file server, and
does not require the file server to keep track of clients. But it also
has its drawbacks.
- I am trying to write
- a program that will enable our users to log even if they are
- trying to access something that has gone stale and therefore
- cause them to hang.
Stale file handles should not cause a hang, since they are definite
responses - although you may not like the response. Unresponsive servers
may cause a hang, look through the mount options for NFS.
- However, none of the files in /etc/ seem
- to reflect a change of status on the filesystem even though it
- has gone stale.
Because it will probably be out of date. Look at your telephone, are there
any stickers: Fred is not in his office. Janice is back from lunch. There
is no point in putting these, since you would reject that information as
stale and phone up Fred or Janice if you wanted something from them,
regardless of any stickers. NFS is very similar.
Thomas
--
*** This is the operative statement, all previous statements are inoperative.
* email: cmaae47 @ ic.ac.uk (Thomas Sippel - Dau) (uk.ac.ic on Janet)
* voice: +44 (1)71 594 6904 (day), or +44 (1)71 594 6957 (fax)
* snail: Imperial College of Science, Technology and Medicine
* The Center for Computing Services, Kensington SW7 2BX, Great Britain
--
*** This is the operative statement, all previous statements are inoperative.
* email: cmaae47 @ ic.ac.uk (Thomas Sippel - Dau) (uk.ac.ic on Janet)
* voice: +44 (1)71 594 6904 (day), or +44 (1)71 594 6957 (fax)
* snail: Imperial College of Science, Technology and Medicine